本研究目的主要在瞭解中部地區二十歲以上高、低利用者之醫療服務利用與醫療服務傾向因素、能力因素及需要因素之關係。本研究以八十四年三月至八月曾至中央健康保險局中區分局特約醫療院所門診就醫一次(界定為低利用者)及就醫五十次以上(界定為高利用者)之二十歲以上保險對象為母群，採系統隨機抽樣法各抽出500人，再從中篩選出在該分局加保且年滿二十歲以上者為樣本，高利用者459人，低利用者399人，以自編問卷為研究工具，進行面對面訪談收集資料，高利用者問卷回收234份，實際分析樣本228人，低利用者問卷回收35份，實際分析樣本134人，共得有效樣本362人。研究資料以SAS/window等套裝軟體進行描述性統計、卡方分析、邏輯迴歸分析等方法進行分析，所得重要結果如下:
一、在「傾向因素」方面：年齡、婚姻、教育程度、職業、健康信念、全民健康保險知識與醫療服務利用具有顯著性相關。高利用者年齡大多為70歲以上者、婚姻為有偶和喪偶者、教育程度大多為不識字者、無工作者、訊息來源以來自醫療人員者多、具外控特質及全民健康保險知識分數中等程度者居多；低利用者年 齡大多分佈於30-29歲、婚姻為有偶和未婚者、教育程度為專科、職業為勞工、具內控特質、訊息來源來自電視者多及全民健康保險知識分數中上程度者居多。
二、在「能力因素」方面：本研究高、低利用者之社會支持程度相差不大，高利用者略高於低利用者，在統計上並無顯著差異；在醫療資源可近性方面，交通時間及就醫總花費時間與醫療服務利用，在統計上具有顯著性差異。
三、在「需要因素」方面：自覺健康狀況、自覺疾病數及自覺症狀數與醫療服務利用，在統計上具有顯著性差異。高利用者其自覺健康狀況較差，自覺疾病數及自覺症狀數較多；低利用者自覺健康狀況良好，自覺疾病數及自覺症狀數較少。
四、傾向因素、能力因素、需要因素等各變項經邏輯迴歸分析後發現其對醫療服務高、低利用的正確預測力為82．3%。其中年齡、婚姻、自覺健康狀況、自覺症狀數、自覺疾病數是影響保險對象醫療服務利用的重要因素。
本研究結果可供保險承辦單位及醫療衛生相關機構在衛生教育宣導時之參考，並提出輔導高利用者適當就醫及後續類似研究之具體建議。

The purpose of the study was to understand the relation of predisposing, enabling, and need factors in health services utilization of higher-users and lower-users over twenty years old in the middle area in Taiwan.
The study extracted data from medical claims field by the contracted hospital and clinics, from March, 1995 to August, 1995, in Central Branch Bureau of National Health Insurance. Subjects were divided into two groups: the higher-users and lower-users group. The higher-users group was the user visiting frequencies over fifty times. The lower-users group was on time.
Data from systematic sampling of 259 higher-users and 299 lower-users. Self-designed close-format questionnaires were utilized. The data were analyzed by frequency distribution, chi-square test, logistic regression. Important findings of study are summarized as follows:
1.Predisposing factors: age, gender, education, marriage, health status, occupation, belief and knowledge of national health insurance were statistical significant difference in the health services utilization.
2.Enabling factors: social support of higher-users and lowers-users were no statistical significant difference; medical availability time of transportation, and total time in health services were significant difference.
3.Need factors: perceived health levels, perceived number of illnesses, perceived symptoms were statistical significant difference.
4.Analyzing predisposing factors, enabling factors, need factors by logistic regression, we found that correct predictive power was 82.3%. The decisive factors that affected utilization of OPD health services were age married status, perceived health status, perceived symptoms, perceived illnesses.
The study is about to provide suggestion to the National Health Insurance related aspects.