全民健保論人計酬試辦計畫影響評估初探─以大臺北地區為例

Abstract

研究目的： 健保局於2011年提出「全民健康保險論人計酬試辦計畫」，分成區域整合模式、社區醫療群模式、醫院忠誠病人模式等3種模式進行試辦，期使醫療服務的提供能轉型以民眾健康為導向，達到促進民眾健康、減少醫療浪費之目的。本研究之目的在評估大臺北地區2個論人計酬試辦計畫模式（區域整合模式、醫院忠誠病人模式）介入後，對醫療資源利用與照護結果之影響。 研究方法： 本研究以大臺北地區（臺北市、新北市、基隆市）之民眾為樣本，利用健保局2010至2012年申報資料進行次級資料分析，採事前事後對照組比較研究，以傾向分數配對（Propensity Score Matching, PSM）法選取對照組，減少選擇性偏差。研究對象為一個忠誠病人模式之醫院，另一為北海岸承作區域整合模式之醫院。區域整合模式以2010年12月設籍於新北市萬里區及石門區且在保之合理現住人口為對照群體；醫院忠誠病人模式以同行政區、同評鑑等級之某區域醫院為對照醫院，並選取2010年全國西醫門診就醫次數於對照醫院為最高（如次數相同則以費用最高），且比率大於等於總西醫門診就醫次數之50%之忠誠病人。其後，使用廣義估計方程式（Generalized estimating equation, GEE），在控制病患特性後，以差異中之差異法（Difference-in-Difference, DID）比較計畫參與者與對照組在醫療利用及照護結果之變化情形。 研究結果： 在醫療利用方面，醫院忠誠病人模式於團隊內門診次數、門診費用、總醫療費用皆較試辦計畫實施前下降，但團隊外門診次數、門診費用、總醫療費用皆較試辦計畫實施前顯著上升；區域整合模式研究組於團隊內門診次數、總醫療費用與團隊外門診費用較試辦計畫實施前上升。在照護結果方面，醫院忠誠病人模式於試辦計畫介入後，急診次數、團隊內院所固定就醫率較介入前為低；區域整合模式團隊內院所固定就醫率較試辦計畫實施前上升。 在樣本配對後進行分析，經控制病患特性，醫院忠誠病人模式研究組於試辦計畫介入後，僅門診次數較對照組為低，區域整合模式各依變項皆未達統計上顯著差異；在照護結果方面，醫院忠誠病人模式研究組於試辦計畫介入後，僅急診次數較對照組為低，區域整合模式各依變項皆未達統計上顯著差異。 結論與建議： 論人計酬試辦計畫實施於短期內確可降低忠誠病患就醫次數，且病患照護結果未因計畫實施有所降低。建議衛生主管機關進行完整計畫評估，並以多重指標定期監測照護結果，以維繫醫療品質。

Background: On July 2011, a 'Capitation Program”was implemented by National Health Insurance (NHI) without registrating with GP required and gatekeeping system, which expected to promote health and downsize health care expenses. Objectives: This study aimed to evaluate the impact of a Capitation Program in Taipei Metropolitan Area on health care utilization and outcomes. Methods: This study adopted the NHI claim data from 2010 to 2012. Patients who joined Capitation Program were recruited as the intervention group and those who never joined Capitation Program were recruited as the comparison group. We used a propensity score matching to select patients in the comparison group to minimize the selection bias. After controlling patients’ characteristics (gender, age, insured category, catastrophic illness or not, Charlson comorbidity index), Generalized Estimating Equation (GEE) were applied to predicted the probability of patients’ enrollment in the Capitation Program. Then we use Difference-in-Difference method to evaluate the impact of the Capitation Program by comparing the health care utilization and outcomes between these two groups. Results: Patients enrolled in the Capitation Program underwent significantly lower physician visits, ambulatory expenses, expenses for all health services and UPC in the model of hospital loyal patient after enrollment. While in the model of area capitation, the physician visits, expenses for all health services and UPC are significantly higher after enrollment. After matching, the health care utilization of patients in the intervention group had a significantly lower number of physician visits in only the model of hospital loyal patient. Concerning the health care outcomes, emergency visits of patients in the intervention group had a significantly lower number in only the loyal patient model while ACSCs is not significant higher. Conclusions and Suggetstions: The program seemed to decline the physician visits of the model of hospital loyal patient, whereas health care outcomes are the same. We suggest that it is necessary for BNHI to evaluate the long term effect on health care utilization outcomes of the Capitation Program. Additional research is required to understand the bussiness strategies of hospital.