Summary: | 博士 === 輔仁大學 === 商學研究所 === 99 === Since the inception of the National Health Insurance (NHI) in 1995, not many literatures of studies on NHI reimbursement for cancer treatment have been reported. Among the literatures published, most studies used the cost items from the Bureau of National Health Insurance (BNHI) database. However, such cost items do not clearly reflect the real cost structure of cancer treatment and therefore further studies are needed for better managerial applications. In this study, we propose a framework to extract useful information from medical records in National Health Insurance Research Database of Taiwan. Two commonly adopted data mining techniques, namely association rules and sequential patternn analysis, are extensively studied to lung cancer patients and their family members in Taiwan. Empirical studies indicated that patient age and gender, disease complication, and hospital ownerships all impose influences of various degrees on consumptions for medical resources. The patients in North and Central District preferred to have counseling and attendance in the same District while the patients in central District tend go to a Integrate Chinese and Western medicine hospital and Chinese medicine clinics. Patients in South District tend to consult other doctors’ opinion before undergoing major treatment. Finally, Patients in Taitung, Hualien, Penghu, and Kinmen County exhibit similar behaviors like those patients in South District.
This study focuses on factors influencing the amount of medical expenses for patients with lung cancer. The research result can provide useful references for the Bureau of National Health Insurance to evaluate its medical expense policies and patient claim compensation standards. It also provides bases for suppliers of medical resources to audit cost control and medical resource allocation mechanism.
|