| Summary: | 碩士 === 明志科技大學 === 工業管理研究所 === 97 === With the economic improvement and changes in demographic structure and lifestyle, the illness suffered by people in Taiwan has also moved from traditional communicable diseases to those of chronic, degenerative nature, and cancer. The illnesses derived from metabolic syndrome (i.e., cardiac, cerebrovascular, diabetic, nephritic, renal, and renal syndrome with high blood pressure) have been occupying the top ten causes of death. These illnesses endangers people’s health and are public health concerns.
This research applied clustering analysis from data mining to group people 40-years and older in Taipei County according to their health risk, using the 2007 integrated community health screening database; the five risk factors in metabolic syndrome are used as attributes. This classification was processed through two stages. The first stage used Word’s method to determine the number of groups, and the second stage compared groups using K-means and EM. CART and CHAID decision tree analysis were then used to identify the primary risk factors that may have caused the metabolic syndrome in these groups.
This study finds that the prevalence of metabolic syndrome among the subject was 30.62%, and the area these people dwelled can be classified into high risk, relatively-high risk, general risk, and low risk areas. The results point out that most high-prevalence area are in remote regions of the county, with the highest (64.56%) being in Wulai Township.
Through examining the characteristic (i.e., dietary habit and lifestyle) of these groups, the researchers have identified some common causes such as smoking, drinking, betel nut chewing, lunching out, and snacking.
This paper contributes to the literature through analysis to assist responsible departments setting appropriate public health policies for a more equitable distribution of healthcare resources.
|
|---|
