Taiwan’s Hospital Distribution and the Access to Emergency Care: A Longitudinal Analysis

研究緣起與目的：過去對於緊急醫療資源研究討論中，多偏向分析急診醫療資源耗用的相關因素，鮮少針對緊急醫療可近性進行探討，並且多偏向單一區域或特定醫療院所為例，使研究的外推性受到限制，同時在國內緊急醫療相關研究上，則缺少關於國人急診醫療資源與急診醫療可近性的長期資料分析探討。 本研究有以下二個研究目的：(1)瞭解臺灣歷年各區提供急診服務醫院家數分布情形；(2)分別以橫斷性資料及長期性資料，檢驗次醫療區急診服務家數對於民眾急診服務利用率與跨區急診率的影響。 在研究對象和資料來源方面，本研究之對象為1997年至2007年，臺灣醫療網分區下之六十三個次醫療區。資料來源為國家衛生研究院發行的2005年100萬人承保抽樣歸人檔。針對不同的目的分析進行方面，(1)描述臺灣歷年各區提供急診服務醫院家數分布情形；(2)針對橫斷性研究部分，採用線性複迴歸模型(Multiple Linear Regression)，探討1997年、2002年、2007年三年，各次醫療區急診家數對於區域跨區急診就醫率及區域急診利用率之影響；針對長期資料設計的部分，採用廣義估計方程式模型(Generalized Estimating Equation Model)探討1997年至2007年次醫療區域提供急診服務的醫院家數與區域內民眾急診利用率、跨區急診率的關係。 研究結果發現：(1)1997年到2007年間，提供急診服務的醫院家數數量維持在一定的範圍內，約230家；提供急診醫療的區域數有增加的現象，急診利用率有逐年上升的情形，但民眾跨區急診率有逐年下降(2)在1997年、2002年、2007年橫斷性研究發現，在控制其他變項後，區域提供急診服務的醫院家數越高，民眾急診利用率越高、跨區急診率越低；在長期資料分析的方面，控制了其他變項後，急診醫院數越多的次醫療區，地區內民眾急診利用率顯著高於急診家數較低的次醫療區(β = 4.237,p<.0001)、整體次醫療區內民眾使用急診服務的比例隨著年代而有顯著提高的 現象(β = 2.866,p<.0001)；急診醫院數越多的次醫療區，地區內民眾跨區急診率顯著低於急診家數較少的次醫療區(β = -2.366,p=0.0001)，整體次醫療區內民眾跨區使用急診服務的比例，隨著年代，而有顯著下降的現象( β = -1.946,p=0.001)；年代與提供急診服務醫院家數交互作用對急診利用率與跨區急診率沒有顯著的作用。 本研究之結論為：臺灣歷年提供急診醫療服務次醫療區域數增多，急診觀察床逐年增加且急診醫院數目維持在一定範圍，民眾歷年急診利用率增加，跨區急診率下降；以長期資料分析後，長期資料分析後，民眾居住地提供急診服務醫院家數越多、以及對照前期的後期年代區域，民眾跨區就醫情形顯著下降，急診利用率顯著上升。整體而言，國人急診可近性增加。建議衛生當局強化現有醫院急診室功能，尤其在資源較不足的偏遠地區或人潮湧擠的觀光地區，可輔導促使急診診療科目增設和資源、人力水準提升，以縮短次醫療區間緊急醫療救護能力差距。Background and Objectives: Previous studies on emergency medical services were about the influcing factors of emergency medicare utilization,few were foucing on the access to emergency care.Partly,the extrapolativeness of past researches has been limited while lacking longitudinal analysis about the changes in Taiwan’s hospital distribution and the access to emergency care at the same time.The purposes of this study were: (1) To Understand the changes in Taiwan’s hospital and emergency department(ED) distribution in 63 sub-region in past decade. (2) To Examine the effect of number of provider toward ED visits and cross-region ED visits,by cross-sectional and longitudinal analysis. Materials and Methods: Subjects between 50 and 70 years-old ,with more than one physicianisit or ED visit in Taiwan were included by using the NHI claim data.We investigate the 63 sub-region area to describe the distribution of emergency hospitals from 1997 to 2007.Data for the analysis came from twenty-five sets of the cohort database from the National Health Insurance research database. Each set included 40,000 persons randomly sampled from the NHI enrollees who were insured during 1997 to 2007. A cross-sectional analysis was conducted to describe the impact of the number of emergency hospital in Taiwan on accessibility of emergency care in 1997,2002 and 2007 by multiple linear regression.Generalized estimating equation model was used to investigate the longitudinal analysis between number of emergency hospital and and ED visit,cross-region ED visit during 1997 to 2007. Results and Conclusions: During 1997-2007，the number of emergency hospitals remain stable in Taiwan.We find there are more sub-area proving emergency care and the ED visits has been increasing,yet the cross-region ED visits has been decreasing in past decade.In 1997,2002,2007,after controlling the other factors,the number of emergency hospitals has strongly positive effect on emergency utilization(β = 4.237,p<.0001),and strongly negative effect on the proportion of cross-region ED visit (β=-2.366,p=0.0001).The’’time’’variable has a strong, positive effect on emergency utilization(β = 2.866,p<.0001) and strong,negative effect on cross-region ED visit( β = -1.946,p=0.001).However,there is no interaction effect between emergency hospital number and year.According to the study result,the health authority should strengthen the fountion of existing emergency hospitals,especially in remote districts ,to make sure the equability in emergency medicare resource