目標：過去之研究顯示失業與醫療利用之關係同時存在健康效應(health effect)與所得效應(income effect)，但迄今尚無定論，在臺灣迄今亦無失業與醫療利用相關之實証研究。本研究之目的在了解臺灣在全民健康保險制度下，失業對醫療利用上之影響，並針對特定疾病醫療利用之影響加以分析。方法：本研究為有對照組之實驗設計，定義之失業為2001年非志願性失業者共92,562人為研究對象，並選取同時期之在職者共92,443人為對照組，觀察期間為每人失業前後各一人年，以差異中之差異法(difference in difference method)與廣義估計方程式(Generalized Estimation Equation )測量。結果：研究結果顯示失業後一年西醫整體門診醫療利用變化呈下降情形，主要受呼吸道感染之醫療利用下降之影響；精神疾病之醫療利用有明顯上升之情形；另三種心血管疾病之醫療利用則在失業後無明顯影響。結論：失業對醫療利用之影響，在不同疾病上確有可能受所得降低與健康衝擊之不同程度之影響。本研究只檢驗失業對醫療利用之短期(一年)影響，至長期影響尚待後續之追蹤與監測。

　　　　　Objective: Previous research has shown that exposure to unemployment could have health effect and income effect on medical care utilization. This study examines total and five disease-specific medical care utilizations before and after exposure to involuntary unemployment under the National Health Insurance Program in Taiwan. Methods: An experimental study with control group and fixed effect model were used. Two groups of people were observed from 1999 to 2002: (1) the involuntary unemployment group (n=92,562), and (2) the control group (n=92,443). The unit of observation was person-year. In order to obtain more precise estimates and control for possible confounders, the difference-in-difference (DD) method and generalized estimation equation (GEE) approach were conducted. Results: Decrease in both ambulatory and inpatient services utilizations were observed after unemployment. In terms of disease-specific utilization, 3 important findings are identified: (1) a decrease in the ambulatory utilization for respiratory diseases has been detected after unemployment; (2) no significant change in utilization for three cardiovascular diseases has been found after unemployment; (3) an increase in utilization has been identified for mental diseases after unemployment. Conclusion: A decrease in income and health effects of unemployment on medical care utilization vary among the types of diseases. One limitation in the study that we only examined the short-term (one year) effects of unemployment on medical care utilization, so it is important in the future to further investigate the possible long-term effects of unemployment on utilization.