台灣各縣市政府醫療支出之影響因素

The factors influencing health care expenditures of Taiwan's city/county governments

彭紹文(Shao-Wen Peng)

中原大學/商學院/國際貿易研究所/碩士(2011年)

https://doi.org/10.6840/cycu201100987

摘要

台灣自實施全民健保以來，國民醫療保健支出幅度有明顯的增加，在各縣市之間因不同的特徵對醫療保健需求會有不同程度的影響，有各種可能因素造成醫療保健支出持續成長，特別是我們強調的各縣市地理區位、都市化程度和都市計畫區現況人口等因素，是否對醫療保健支出產生影響是值得研究探討的。

本研究採用1998 至 2009 年台灣23縣市縱橫資料進行探討，同時用固定效果模型與隨機效果模型進行評估，並用 Hausman 檢定選出較佳模型，最後，再用Wald檢定進行差異性斜率分析，以評估各因素對醫療支出的影響是否隨各縣市政府不同而有差異。首先，探討各縣市政府醫療支出影響因素所顯示差異性資料。其次，將23縣市區分為不同的地理區域群體來評估各不同群體對醫療支出的影響。第三，研究的重點在於評估各縣市都市化程度現況人口對政府醫療支出的影響。最後，利用差異性斜率的估計方法，以更精確地觀察各縣市中影響醫療支出的因素之內涵。

由本文實證結果可歸納出幾個重點，第一、台灣地區全體23縣市中，每萬人口執業醫事人員及醫療保健支出物價指數是影響醫療保健支出的主要因素。第二、以地理區域而言，隨著地理區位不同，各變數對於醫療支出的影響，明顯存在差異性效果。因此，為了提升醫療政策的效率，應考量地理區域的特性與差異性。第三、在都市化程度較高及人口數較多之縣市中，以接受急難救助平均每人次金額、失業率及醫療保健支出物價指數等因素對於醫療支出具有顯著影響，建議未來醫療政策亦須針對縣市都市化程度不同進行考量，避免城鄉醫療品質差距懸殊。

關鍵字

差異性斜率 ； 縱橫資料 ； Wald檢定 ； Hausman 檢定

並列摘要

Since the National Health Insurance Program was implemented in Taiwan in 1995, the national health expenditures have obviously increased. The local county/city governments of Taiwan have their own specific characteristics; therefore, these characteristics may give rise to different effects on healthcare. This study estimates the effects of important factors on local governments’ medical care. We highlight the factors such as geographical location, urbanization, and present population of urban planning districts.
To verify the above augments this study employs the panel data of Taiwan’s 23 counties/cities during 1998-2009. The estimation models are fixed effects model and random effects model, and the optimal model is chosen employing the Hasuman test. Moreover, we use the Wald test to evaluate whether the effects of each variable on healthcare expenditures are different among counties/cities governments. In empirical estimation, this research further divides 23 counties/cities into different groups according to geographical locations to evaluate the differential effects on healthcare. In addition, we also investigate whether the urbanization influence healthcare.
The empirical estimations have the following findings: First, in the entire 23 counties/cities of Taiwan registered medical personnel per 10000 population and the price index of healthcare are the major sources influencing the expenditures of healthcare. Second, the effects of each variable on medical care vary from geography to geography. Thus, to improve the efficiency of medical care policy the consideration of geographical difference is necessary. Third, average amount of the disaster assistance received per capita, unemployment rate, and the price index of healthcare are the significant factors influencing the expenditures of healthcare for the counties/cities with high urbanization and population. Thus, to avoid the enormous disparity in healthcare quality the healthcare policy also needs to consider the difference of urbanization in each county/city.

並列關鍵字

Panel data model ； Hausman test ； Wald test ； differential slope analysis

參考文獻

邱雅苓、陳世能 (2001)，醫療保健支出成長因素之探討－時間序列與門檻模型的應用，

邱雅苓、陳世能 (2003)，醫療保健支出成長因素之探討─時間序列分析與門檻模型的應

游慧光、洪乙禎、王漢民、謝啟瑞 (2007)，醫療保健支出的跨國與跨時比較，社會科學

鄭新慶 (2004)，台灣醫療支出成長因素之探討--時間序列與有向圖之應用，逢甲大學會

張正蓉 (2006) ，台灣及OECD國家醫療體系治理、經濟發展、醫療支出、醫療資源、