由於不同醫院權屬別有不同的目標函數和預算寬裕度，故不同醫院權屬別的財務誘因及行為也會有所不同，因此，本文旨在探討醫院權屬別與病患的醫療結果和醫療利用的關聯性，並以第三、四期女性乳癌病患做為研究對象。本文使用2007-2012年全民健康保險資料庫和癌症登記資料進行實證研究，並藉由癌症登記檔選擇初次確診罹患第三、四期乳癌病患當作研究對象。我們以病患初次罹癌後的1年內、3年內死亡率和1年內、3年內醫療點數，做為衡量病人醫療救治結果的指標。在實證分析時，由於病患可能依據本身疾病嚴重程度來選擇醫院，進而產生樣本自我選擇的問題，因此本文以兩階段最小平方法做為實證模型，工具變數使用病患居住地30公里內醫學中心家數（不論權屬別）、病患居住地30公里內區域等級以上的公立醫院家數、病患居住地30公里內區域等級以上的財法醫院家數、最近宗教型財法醫學中心距離、住家鄉鎮附近是否有醫學院校的虛擬變數等。 本文的實證結果得到，和宗教型財團法人醫院相比，在醫療結果方面，公立醫院對於3年內死亡率有負顯著的影響，顯示在公立醫院做長時間治療的末期病人，其所獲得的治療結果比宗教型財團法人醫院較不佳；在醫療利用方面，公立醫院或一般型財團法人醫院，對於病患1年內和3年內醫療點數皆有負顯著的影響，顯示病患的醫療利用會受到醫院權屬別影響。;Due to the difference in objective function of hospital ownerships, it turns out different financial incentives and managements. This paper aims to analysis the relationship among hospital owners, patients’ treatment outcomes and medical utilizations. We concluded the third and the forth stage breast cancer patients in this sample. The data was from Taiwan National Health Insurance and Taiwan Cancer Registry between 2007 and 2012. We control patients’cancer stage, then we used one-year and three-year mortality rates to measure the patients’ treatment outcomes. Additionally, we adopted one-year and three-year medical dot to estimate the patients’medical utilization. Concerning that patients might choose admitted hospital depend on their illness severity, we utilized the two-stage least square model to diminish the sample self-selection bias. The five instruments used in this paper are as following: the numbers of medical center of patients’ house within 30 kilometers, the numbers of regional public hospitals of patients’ house within 30 kilometers, the numbers of regional nonprofit of patients’ house within 30 kilometers, the shortest distance between religious nonprofit medical center and the patient’s house, and the dummy variable which represents if there is any medical colleges in town. The result shows that, compared to the religious nonprofit hospital, the public hospital has a positive significant influence on three-year mortality rate. Furthermore, the public hospital and the general nonprofit hospital have the negative significant influence on one-year and three-year dot. These results prove that the treatment outcomes and medical utilization are affected by hospital ownership.
