總額預算制度對急診醫療資源利用及品質之影響

目的：本研究探討健保局高屏分局實施醫院總額急診醫療照護管理方案後，各級醫院各年度之急診就醫趨勢及醫療照護品質變化。研究方法：以高屏分局特約申報急診121家地區級以上醫院（3家醫學中心、10家區域醫院及108家地區醫院）急診就醫之個案為研究對象。研究資料為2004-2008年期間之申報費用，並依總額預算制度施行階段分為三期：總額初期(2004-2005年)、總額成長期(2006-2007年)、總額成熟期(2008年)。研究期間符合「急診就醫」條件者計有5,605,852人次，各年度人次則在1,085,110（2008年）至1,169,381（2005年）之間。分析方法有：卡方檢定（類別變項）、變異數分析（連續變項），並利用邏輯斯迴歸探討個別因子之影響結果。研究結果：在急診醫療資源配置方面，不同年度之間，各層級醫院家數及急診暫留床數無顯著性影響，急診專科醫師數則有顯著差異(p<0.001)。在病患人口學方面：男性(54%)多於女性(45%) (p<0.001)；19至44歲就醫人數最多(33%)、其次為45至65歲及0至18歲(分別為22%)(p<0.001)；「投保類別」以第一類投保人口最多(36.9%)。在急診醫療利用方面，以區域醫院收治人次占率最高(36%)，其次為地區醫院(29%)(p<0.001)；疾病嚴重度以檢傷分類第三級、Charlson index 0及1分、而疾病分類以損傷及中毒、呼吸道疾病分占最高(19 %) (p<0.001)。在急診醫療品質方面，正向指標之「急診檢傷第1、2級轉住院率」呈現逐年成長(p<0.001)；負向指標之「3日內急診返診率」則是2008年較2007年減少，醫學中心「滯留急診暫留床比率」從2007年開始呈現成長(p<0.001)，死亡率則是逐年減少(p<0.001)。結論：在急診醫療資源配置方面，地區醫院家數隨年度減少，但在整體急診醫師人力與急診暫留床數皆有成長情況下，顯示並未影響整體醫療供給。在急診就醫趨勢方面，轄區大型醫院整體急診收治人次呈現明顯上升，且檢傷分類第三、四級之人次與Charlson index 0及1分佔率成長，將使轄區醫院急診醫療照護量之負荷愈形加重。在急診醫療品質方面，整體而言，轄區急診返診率、急診死亡率呈現下降趨勢，但「檢傷1-2級轉急診住院佔急診住院病患比率」呈現上升趨勢，顯示醫院未有拒絕收治嚴重度較高病患現象。醫學中心之「急診滯留2日病患比例」於2007年後明顯上升，尤以檢傷分類第三級未達住院條件病患最多，可見病患需適量移轉至非醫學中心層級就醫，以利解決醫學中心急診滯留增加之情況。建議：由結果得知，急診壅塞之部份原因為初級照護病人未能由適當醫療機構收治，故保險人應設計相關配套措施，達到醫療資源合理分配以紓解急診壅塞情形，並提升急診之照護品質。建議後續研究者以全國性資料進行長期性的時間序列分析評估；此外，以問卷方式併行調查，探討需求面就醫行為的影響，也可針對供給面深入探討醫院行為之影響。

關鍵字

醫院總額；急診醫療照護管理方案；急診醫療利用；急診醫療品質

並列摘要

Objective: This study aimed to analyze the trends of emergency visits and emergency care quality changes in different levels of hospitals after the implementation of the global budget payment system by the bureau of National Health Insurance, Kao-Ping Branch. Materials and Methods: Included for analysis were health expenditures of patients with emergency visits to the 121 hospitals (three medical centers, 10 district hospitals and 108 regional hospitals) contracting with the bureau. The study period 2004-2008 was divided into three phases: the introduction stage (2004-2005), the growth stage (2006-2007), and the maturity stage (2008). In total, there were 5,605,852 patients who met the criteria to be included for analysis during the study period, with the annual numbers of emergency visits ranging from 1,085,110 (the lowest, in the year 2008) to 1,169,381visits (the highest, in the year 2005). The Chi-square, ANOVA, and multiple logistic regression techniques were used for statistical analyses. Results: In terms of medical resources, there was no significant difference across the three phases in the emergency beds, but the numbers of emergency physicians differed significantly (p<0.001). Most patients were in the range of 19 to 44 years old (33%), followed by the ranges of 0 to 18 and 45 to 65 years old (22%, respectively) (p<0.001). More than one third (36.9%) of emergency service users were category one insurees of the national health insurance program. Regarding the utilization of medical resources, the district hospitals had most of the patients (36%), followed by the regional hospitals (29%) (p<0.001). Most patients had a disease severity of triage III, Charlson index 0 and 1, trauma and toxication and respiratory disease (19%) (p<0.001). As for the emergency medical quality, the positive index of “transfer of patients with emergency triage I and II” increased yearly. The negative index of “returning rate within 3 days” was lower in 2008 than in 2007, the “patients’ stasis rate in medical centers” increased from 2007, and the mortality rate decreased year by year (p<0.001). Conclusions: In the aspect of medical resources, although the number of district hospitals decreased, the number of emergency beds increased. Therefore, no significant impact on the supply of medical services was observed. The trend analysis showed that emergency visits to larger hospitals increased and patients with triages III and IV and Carlson index 0 and 1 increased as well. This would increase the burden of the emergency department. About the quality of emergency service, overall the returning rate and mortality rate decreased. But the transferal rate for patients with triage I and II on the admission list increased with no refused care of critical patients. The “stasis rate with more than 2 days in the emergency department” of medical centers had increased from 2007, especially among patients with triage III with no need for admission. Therefore, a better mechanism to transfer patients to appropriate hospitals rather than medical centers might resolve the stasis problem for the medical centers. Suggestions: The primary care patients not properly distributed and treated in suitable hospitals is one of the reasons for the crowding in the emergency department. A better design of transferal system might help solve the crowding problem. Further researchers might consider using national database to conduct time-series analyses for a long period of time. Also, a questionnaire survey might help identify the need of patients with emergency visit and delineate the behavior of hospitals providing emergency medical service.