[[alternative]]The possible impact of outbreak of pandemic H5N1 influenza medical utilization- experience from SARS

[[abstract]]背景：若一旦爆發禽流感大流行，將可能造成對醫療體系的衝擊甚鉅，然而我國目前尚未有禽流感人傳人疫情爆發所需之醫療利用的研究。因此，本研究之結果可以提供政府相關部門研擬因應禽流感大流行相關措施之參考或依據。目的：本研究旨在以嚴重急性呼吸道症候群（SARS）相關病患的醫療利用資料，推估禽流感人傳人疫情爆發時，在不同侵襲率下的住院費用總額及探討與住院醫療利用和病患跨區就醫相關的因素。方法：本研究資料是運用國家衛生研究院發行之全民健康保險資料庫，選取與禽流感傳染毒力相似，而住院診斷碼為V01.8、079.89、480.8及518.82的366位SARS相關住院病患作為本研究之研究對象，且以行政院衛生署疾病管制局所公佈的「因應流感大流行執行策略計畫」中預測不同侵襲率下的住院人數資料作為費用推估之依據。採用SAS 9.1及SPSS 15.0統計套裝軟體進行資料分析，統計方法包括卡方檢定、獨立t檢定、變異數分析、複迴歸及羅吉斯迴歸分析等。結果：在住院費用總額推估方面，侵襲率為15%、25%、35%時，只考量病患年齡時的估計值分別約為4.28億元、7.14億元、9.99億元；同時考量醫院層級別與病患年齡時的估計值分別約為11.05億元、19.06億元、25.79億元；同時考量是否院內死亡與病患年齡時的估計值分別約為11.69億元、19.48億元、27.27億元。其中以同時考量是否院內死亡與病患年齡時所推估的住院費用總額最高，且以65歲（含）以上的老年人相較其他年齡層病患的總住院費用高。其次，在與住院費用相關的因素方面，獨立影響住院費用的顯著相關因素有病患性別、是否院內死亡、醫院層級別、醫院權屬別及住院天數。男性較女性的住院費用高29,571元；院內死亡者較非院內死亡者的住院費用高151,603元；區域醫院相較醫學中心的住院費用低61,355元，而地區醫院相較醫學中心的住院費用高37,979元；私立醫院與財團法人醫院相較於公立醫院的住院費用高，分別高66,205元及48978元；住院天數與住院費用呈正相關，住院天數每增加一天時，住院費用則會增加5,948元。此外，在與住院天數相關的因素方面，獨立影響住院天數的顯著相關因素為就醫醫院層級別。地區醫院較醫學中心的住院天數明顯長26.65天，且對住院天數達統計上顯著水準。最後，在與跨區就醫相關的因素方面，獨立影響跨區就醫的顯著相關因素有疫情嚴重度與每萬人口急性一般病床數。輕度疫情區的跨區就醫勝算較嚴重疫情區明顯來得高，其勝算比為3.61；每萬人口急性一般病床數的勝算比為0.93，表示每增加一床則跨區就醫勝算會減少7%。結論：65歲（含）以上的老年人較其他年齡層病患的總住院費用高，因此建議衛生主管機關事先制定相關衛生政策，例如：疫苗的施打、藥物預防或增加相關宣導等，以降低老年人罹病之風險及減少醫療費用的支出。其次，在就醫醫院層級別方面，以地區醫院的住院費用最高、區域醫院則最低，因此建議衛生主管機關在規劃禽流感專責醫院時應以區域級醫院為優先考量。關鍵字：禽流感、醫療利用、嚴重急性呼吸道症候群、跨區就醫[[abstract]]Background: The spread of the H5N1 influenza pandemic is expected to tremendously impact the healthcare delivery system. However, no studies have evaluated the medical resources and expenditures required in the event of an outbreak. The findings of study may help government agencies design related programs.Objectives: This study attempts to determine the admission expenditures required for responding to different attack rates of pandemic H5N1 influenza and explores the factors related to medical utilization and cross-region admission based on the experience of Severe Acute Respiratory Syndrome (SARS) outbreak. Methods: Data were retrieved from National Health Insurance (NHI) database developed by National Health Research Institution (NHRI) among inpatients with SARS in 2003. Cases classified as V01.8,079.89, 480.8 and 518.82 by International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) were retrieved for analysis. Following the descriptive data, the researcher adopted the inferential statistical techniques, including independent t-test, χ2-test, ANOVA, linear regression and logistic regression.Results: The findings revealed that, for attack rates of 15%, 25% and 35%, estimated admission expenditures were 428,139,118 NT dollars, 713,552,389 NT dollars and 998,974,253 NT dollars, respectively, when considering only patient age. Estimated admission expenditures were about 1,105,073,348 NT dollars, 1,905,589,661 NT dollars and 2,578,478,074 NT dollars when considering both hospital level and patient age . Estimated admission expenditures were about 1,168,737,578 NT dollars, 1,947,861,142 NT dollars and 2,727,023,898 NT dollars when considering in-hospital death and patient age simultaneously. Total admission expenditures for elderly patients over the age of 65 years were higher than those for other age groups. Secondly, factors related to admission expenditures were patient gender, in-hospital death, hospital level, hospital ownership and length of stay (LOS). Besides, the related factors of LOS were hospital level. The LOS in local hospitals was 26.65 days longer than that at medical centers. Finally, factors related to cross-region admission were severity of the epidemic and bed-population ratios.Conclusions: Total admission expenditures for elderly patients over the age of 65 years were higher than those for other age groups. Hence, we suggest the specific health policy for this group such as vaccination, drug prevention or a public awareness campaign to reduce the risk of disease and medical expenses. Secondly, because admission expenditures in local hospitals were the highest and regional hospitals were the lowest, we suggest that regional hospitals be given priority when health authorities plan exclusive hospitals for treating H5N1 influenza.Key words: H5N1 influenza, medical utilization, Severe Acute Respiratory Syndrome, cross-region admissio