醫院經營效率與領導型態關聯性探討：以國軍醫院為例
The Relationship Between Hospitals＇ Efficiency and Leadership
Style：Examples of the Military Hospitals
中文摘要 本研究藉由資料包絡分析法評估國軍醫院相對經營效率，瞭解各醫院在體系內相 對經營效率的程度，並可藉由差額分析結果建議重新調整投入與產出的資源分 配。在組織績效的認定上，領導績效亦包含於組織績效內，相信領導行為是一個 影響組織績效的重要變項，有些學者認為對提高組織績效而言，主管的領導型態 比其他的管理因素更為重要，故領導型態之研究，也就與績效有關聯性存在。 本研究以經營效率為依變項，領導型態為自變項，探討二者間之關聯性。經營效 率之次級資料收集 15 家國軍醫院 88 至 92 年度由管理機關所提供服務量資料， 為主要研究對象。院長領導行為之初級資料收集，則採用 LBDQ XII 量表其中四 類構面題項，加入個人主觀認知組合而成之結構性問卷，經信效度測試後，由醫 院之一級主管（含任務編組）填答來描述院長領導型態。 經由資料包絡法（DEA）衡量國軍醫院 5 年度計 75 個決策單位（DMUs）值， 相對有效率的醫院計有 32 家（E=1），相對無效率的醫院計有 43 家。領導行為 問卷發出 510 份，回收問卷 346 份，有效問卷 282 份，有效問卷回收率為 55.29 ％。15 家國軍醫院 5 年度期間，院長領導行為經因素分析後重新定義歸屬為「高 倡導高體恤導向領導」、「高倡導導向領導」及「高體恤導向領導」等三種領導型 態，其中以「高體恤導向領導」型態最多。 依據醫院評鑑等級與醫院規模分組來檢定，不同醫院評鑑等級與不同醫院規模 間，研究發現並無顯著差異，另外相對有效率醫院與相對無效率醫院之投入變項 間與產出變項間，亦無顯著差異，其可能原因為：一、雖然部分文獻建議希望能 將有關財務面之資料放入 DEA 模式中運算，但由本研究實際操作後發現，加入 財務面資料之評估後，國軍醫院的相對效率值皆提昇，意即降低 DEA 模式的區 辨能力。二、公立醫院皆有不同程度的經費補助，無法真實顯現實際成本，未經 調整或還原的成本數據加入 DEA 模式評估，將會降低 DEA 模式的區辨能力， 建議僅選取醫療收入放入 DEA 模式評估。 對於院長領導型態與醫院經營相對效率關聯性分析結果，在分別控制醫院評鑑等 級與醫院規模變項後，以 Tobit 迴歸模式分析檢定後，皆與醫院經營相對效率無 顯著相關。可能原因為：一、國軍醫院屬於公立醫療體系之一環，內部受到人事、 採購、預算、會計、審計等相關法令及政策規範下，在經營管理上受到相當大的 限制，缺乏彈性及迅速反應能力，限縮了醫院院長領導醫院的幅度。二、醫院的 收入絕大部分來自醫師執行醫療作業產生醫療收入，如果院長的領導型態並沒有
介入醫師的獎酬制度，醫師較以自我為中心，因而院長領導型態與經營效率無顯 著相關。 國軍醫院在制度面可朝向法規鬆綁，流程再造與多元化經營努力，各醫院在實際 經營面對於投入（醫師數、病床數、總成本）/產出（醫療收入、門急診人次、 總住院人日）變項尚有減少與增加的幅度，可供院長經營管理努力。國軍醫院院 長領導績效可能侷限於法令規章的桎梏無法彰顯，可以主動爭取領導管理的幅 度，創造領導績效。 英文摘要
In this research, the Data Envelopment Analysis has been applied to evaluate the relative efficiency of military hospitals in Taiwan. By investigating the difference of relative efficiency of each individual hospital’s, we hope to precisely modify the current input-output resource allocation and thus utilize resources more efficiently. The performance of leadership is part of the performance of organization that it acts as a critical variable. Some scholars actually believe that the style of leadership is the most important criterion in enhancing the efficiency of the organization. Therefore, the analysis of leadership style is valuable to the performance.
What we are interested in are finding out the relationship, if there is any, between performance efficiency and style of leadership. This paper is organized as follows: First, DEA is conducted to evaluate the relative performance efficiency of different military hospitals. Next, we attempt to exam the relationship between efficiency and leadership in an in-depth way by adopting Tobit regression model. The performance efficiency is the dependant variable in our model, while the style of leadership is the independent variable.
Our secondary data, Year 88-92 annual report on medical service data are collected from the administration departments of 15 military hospitals. On the other hand, the four aspect items and the structural survey from the LBDQ XII are used as our primary data. Notice that the survey is structured with subjective opinion and answered by hospital’s top managers (include task-unit) who offered literal description of the leadership they have.
The DMUs of the military hospitals’ 75 decision-making units are measured by DEA. We found that there are 32 relatively efficient hospitals and 43 relatively inefficient hospitals. 510 surveys regarding the leadership behavior had been sent out and 346 had been returned. There are 282 valid surveys that the response rate is 55.29%. According to the survey by the factor analysis, the 75 DMUs leadership style has been defined as a three-factor construct composed of “high initiating structure and high consideration”, “high initiating structure” and “high consideration”. Among them, the group of “high consideration” is the largest set which takes 58.33%.
When we examined hospitals according to their ranks and sizes, we found that there is no significant difference between them. On the other hand, while we compared
relatively efficient hospitals and relatively inefficient there is no meaningful
difference between input and output variables. The possible underlying reasons are as follows: First, some literatures suggest that we should put financial data into the DEA. After this analysis, we discover that adding financial data would raise the value of relative efficiency.and lower the distinguish power of DEA.On the other hand, since public hospitals receive different amount of financial aid from government that it is not possible to obtain the realistic input cost. The DEA model distinguishing would be lowered if we put not any adjustment cost into DEA output variable, therefore we only considered medical revenue as the input variable.
Regarding to the relationship between leadership style and hospital’s efficiency, after we control hospital ranks and size and investigate it by Tobit regression , we discover that they are not critical to the relatively efficiency of hospital management. The possible underlying reasons II are as follows: first, since public hospital is under the public medical system that their management is restricted by the factors such as personnel system, public purchase, budget, finance and audit. There is lack of adjustment and prompt response ability that it limits the range of the leadership. Secondly, the most part of doctors’ salary come from giving out medical treatment to their patients that if the leadership of the hospital is not involved in doctors’ reward system then doctors do not care much about the leadership they face.
