Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Thomas C. Nowak, Ph.D.

Second Advisor

D. Alex Heckert, Ph.D.

Third Advisor

John A. Anderson, Ph.D.


After the implementation of National Health Insurance (NHI) program in 1995, all residents in Taiwan became universally insured and had the right of access to the health care system. Nevertheless, according to statistics from Bureau of National Health Insurance (2004), the expenditures of the NHI have exceeded its revenue since 1998. Despite the persistent public debate on the fiscal conditions of NHI, little analysis has been done to identify the individual factors that cause rising expenditures. A key component of many recent health reform efforts is a system of risk-adjusted payments among health plans (Newhouse, 1998; Iezzoni, 2003). To provide a complete and balanced assessment of the NHI, it is necessary and pertinent to assess the consumers’ medical utilizations in the NHI program. The purpose of this study is to identify the factors (i.e. demographic factors, diagnostic factors, etc.) that may explain rising expenditures in NHI program. Furthermore, this study will try to develop a risk adjustment model based on diagnostic and demographic information from NHI dataset. The risk adjusters will then be evaluated for their predictability. This study was conducted as a secondary data analysis. All of the data used in this study were obtained from the National Health Insurance Research Database (NHIRD) provided by National Health Research Institutes in Taiwan. All of the individuals in the NHIRD are re-coded therefore they are anonymous. This study will first develop a Taiwanese version diagnostic base risk adjustment model by using one of the most popular risk adjustment software on the market-diagnostic cost groups (DxCG) (Ellis & Ash, 1995; Iezzoni, 2003; Richardson, 2006). Second, by utilizing multiple regression and logistic regression, risk factors will be identified. Finally, ROC (receiver operating characteristic) curve analysis then will be used to evaluate the predictability of the risk factors. The results of this study may help produce health care reform for Taiwan toward a better system, as well as provide knowledge of Taiwan’s health sector achievements and NHI experience for scholars and policy experts in the United States. Taiwan’s Bureau of National Health Insurance (BNHI) may utilize this information to formulate a new payment system, with a focus on efforts to control health cost inflation.