| Summary: | 碩士 === 中原大學 === 工業與系統工程研究所 === 104 === This research aims to model the lifetime cost of coronary heart disease (CHD) for people with 3H (hypertension, hyperglycemia, and hyperlipemia). The purpose of this study is to study the burden on CHD patients and National Health Insurance Bureau the national insurance medical cost caused by 3H.
We use the NHIRD (National Health Insurance Research Database) data from years 1999 to 2008 for model fitting and cost analysis. The International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes used are: 250.xx for hyperglycemia, 401.xx-405.xx for hypertension, 272.xx for hyperlipemia and 410.xx-414.xx for CHD. The first 3 years (1999 – 2001) are set as the observation period. We select the patient data recoded in these 3 years but without 3H or CHD symptoms record, and delete data without complete information. The total population is 609,944. The next 7 years (2002 – 2008) are set as the experimental period. The patients with 3H symptoms in the experimental period are set as the case group. We then use matched-pair method to select patients without 3H, called the control group, by matching the gender exactly, birth date closely, and first-visit date closely of the patients in the case group.
The methodology of this study consists of two parts: estimate the CHD incidence rate and calculate the CHD lifetime cost. We estimate the CHD incidence rate by dividing the number of CHD cases by the person-years of all cases. To calculate the CHD lifetime costs, we modify the method by Taylor et al. (1996) and use the CHD survival rate by Proudfit et al. (1983). The CHD lifetime cost includes the direct medical cost and indirect cost. The direct medical cost includes the ward fees, physician medical fees, drug costs, rehabilitation costs, imaging inspection fees, laboratory fees, registration fees and others. The indirect cost includes the loss due to patients’ lost productivity during outpatient and inpatient medication period. Only the outpatient and inpatient medical costs for CHD are considered in this research.
The results show that the overall CHD incidence rates for 3H and non-3H groups are 3.5% and 0.98%, respectively. The average number of ambulatory-care days and hospital days for the 3H group per person are 73.52 days and 13.2 days, respectively, and those for the non-3H groups per person are 76.93 days and 12.22 days, respectively.
If we consider only cases having CHD during the experimental period, the average CHD lifetime costs per person for the 3H and non-3H groups are 102,831 and 96,918 dollars, respectively. Since the mediccation costs for the two groups do not differ much, the CHD lifetime costs for the two groups are similar.
If we consider all cases in the two groups, the average CHD lifetime costs per person for the 3H and non-3H groups are 12,333 and 5,480 dollars, respectively. This result shows that the 3H group with a higher CHD incidence rate causes a larger burden on the society medical cost.
Finally, we conduct a sensitivity analysis for the present-value discount rate and coronary heart disease survival rate. When the present-value discount rate increases from 2% to 3%, the CHD lifetime cost decreases 0.015% and 0.019% approximately for the 3H and non-3H groups, respectively. When the survival rate decreases 30%, the CHD lifetime cost also decreases approximately 30% from its original value for both groups.
|
|---|
