中央健保局於92年開始推行「家庭醫師整合性照護制度」試辦計畫，其主要弁酮偌ㄗ悒螻釧P全性、協調性與持續性的初級醫療與預防保健服務，並以民眾健康為導向，建立以病人為中心的醫療觀念，提升醫療服務品質。不過，由於民眾長久養成的就醫習慣以及對基層診所看病的疑慮與不信任，使得家庭醫師制度的參與一直到現在都還不普及。為了找出影響民眾參與家醫計劃的重要因素，本研究利用94年「南區區域家庭醫師整合照護計劃管理及探討」之病人問卷調查資料，以資料探勘技術之決策樹演算法進行分析，找出影響民眾參與家庭醫師整合照護制度之行為規則模式。研究結果發現：是否知道家庭醫師整合試辦計劃、教育程度、知道計劃的消息來源等因素，為造成影響民眾參與計劃的重要因素；知道家庭醫師試辦計劃之民眾和得知該計劃是透過診所醫師的民眾，較易加入家醫會員。教育程度為初中以上之民眾和透過傳播媒體(報紙、雜誌、電視、廣播)得知計劃的民眾，較不會加入家醫會員。建議應促進醫病雙方互信的關係，使民眾與基層醫師產生相互依賴與信任，加強家庭醫師制度方面的教育宣導，使其瞭解實施家庭醫師制度的重要性，才能有效導正病人的就醫行為。 The trial plan of family physician integrated care system has implemented by Bureau of National Health Insurance in 2003. Purposes of this plan are providing comprehensive, coherent and continuing primary health care and preventive services, and to establish a patient-orientated concept to enhance service quality of health care. The participative rate is low because of patients’ medical seeking behaviors and patients’ doubt and distrust in the primary health care.
In order to find out the significant factors which influence patients’ participation in this plan, this research used decision tree of data miming techniques to develop a behavior model by analyzing a database on patient's questionnaire survey of southern district regional family physician planned management in 2005. Results of study found, that whether knowing the family physician integration trial plan or not, education, and information sources are important factors to influence people participating in this plan. Members participated in the family physician intergraded trial plan knew this program well and got the information from primary physicians. Subjects educated above junior high school and got the plan information from mass media would not join to be members. Results suggested the relationship between physicians and patients should be promoted to build patients faith in primary care physicians, education of the family physicians’ system should be emphasized to improve patients’ understanding of importance of the family physician to guide patients’ medical behavior effectively.
