以資料包絡法探討內外科護理單位之工作效率

近年來由於醫療環境變化迅速，健保支付制度已由論量計酬改為總額預算制度，醫院為了提升醫療市場之競爭力，越來越重視醫院績效管理，因此醫院管理者需追求醫院經營績效的極大化，以發揮醫院的經營效率，而護理單位是醫院經營最基本之單位，其投入的資源與產出的結果，對醫院經營具有很大之影響。但探討護理單位的整體效率之研究較缺乏，因此本研究運用資料包絡法（Data Envelopment Analysis, DEA）模式，探討南部某醫學中心內科組15個護理單位與外科組13個護理單位之工作效率，內科組選取護理人員數、護理時數、教育訓練時數與工作量指數為投入變項，住院人日、住院人次及佔床率為產出變項。外科組選取護理人員數、護理時數與工作量指數為投入變項，住院人日、住院人次及手術人數為產出變項。利用2006年8月1日至2007年7月31日實際臨床資料，以每一個護理單位為一決策單位（Decision-Making Units, DMU），衡量其整體效率（Overall Efficiency）、純技術效率（Pure Technical Efficiency）、規模效率（Scale Efficiency），找出參考組合（Efficient Reference Set, ERS）作為相對無效率的護理單位參考，並利用差額變數分析(Slack Variable Analysis, SVA)提供相對無效率之護理單位各投入及產出項改善方向與幅度，以提昇工作效率。分析結果得知，一、整體無效率之護理單位內科組有7個（46.67%）、外科組有6個（46.15%）。二、純技術無效率之護理單位內科組有4個（26.67%），外科組有5個（38.46%）。三、規模無效率之護理單位內科組有7個（46.67%）、外科組有6個（46.15%）。相對無效率之護理單位，其效率值介於.9-1之間分別有5個及3個單位，屬邊緣非效率單位，只需在投入項稍作調整即可達到相對有效率。效率值＜.9共有5個護理單位，屬於明顯非效率單位。依據差額變數分析針對相對無效率內、外科組，須減少投入變項之資源為血液腫瘤科、腦神經科、心胸腔外科、腫瘤外科與骨口外等護理單位，除骨、口外科歷經兩次單位合併，床位利用率不佳之外，其他四個護理單位有其特殊性，病人病情複雜，護理依賴度高，須較多之護理人力。外科組護理單位之病人，因為手術前後所需護理照護需求較多，所以應該依當日手術病人數，安排適當人力，因此在考量規模效率下，要達成整體有效率，投入與產出項之調整不限於投入必須減少，而是視各單位之情況做投入與產出之增減。護理工作是專業性工作，知識須不斷更新，以創新服務，因此積極參與在職教育勢必需要，而護理人員上班是採輪班制，所以應繼續再發展多元化彈性的學習方式，如在各單位裝置PowerCam，隨時錄製在職教育內容做成DVD，或採用數位學習同步或非同步進行自我學習，以培養護理人員的專業角色與成長。另外為配合醫院病床流用政策，護理人員必須跨科交叉輪調訓練學習，才能達到人員彈性應用。所以醫院在評估護理人力除考慮病人嚴重度外，尚需做護理專業與非專業（如：護佐）工作之重組，以部份比例之護佐參予部份比例之護理人力（Skill-Mixed Model）執行臨床護理工作，以增加病人所獲得之護理時數及護理人員之工作滿意度，這前提之下，這些護佐需接受臨床照護技術層面之專業、正確性訓練（例如：抽痰技術訓練等）。

關鍵字

護理單位；資料包絡法；相對效率

並列摘要

Due to rapid change of medical environment in the recent years, the payment of health insurance was changed from case payment to global budget. To increasing the competitiveness in the medical market, the performance and management of hospital get more and more attention. Therefore, the manager of hospital should expand the profit and pursuit the most effective performance of hospital. The nurse’s station is the base unit of hospital. The result of resources input has a great impact in the hospital management. But there is lack of researches about the effective of nurse’s station. Then, this research used Data Envelopment Analysis (DEA) to explore fifty medical nurse’s stations and thirteen surgical nurse’s stations. The input variants of medical nurse’s station were nurse’s numbers, hours of nursing, hours of education, loading of work, and the output variants were length of hospital stay per patient, number of patient, the rate of bed occupation. The surgical units used number of nurse, hours of nursing, and loading of work as the input variants, and length of hospital stay per patient, number of patient, number of operation as the output variants. The data was collected from August 1, 2006 to July 31, 2007. Each nurse’s station was used as Decision-Making Units (DMU) and the Overall Efficiency, Pure Efficiency and Scale Efficiency were evaluated. Thereafter, we obtained the Efficient Reference Set (ERS) as the reference of the Relative Inefficiency nurse’s stations. The Slack Variable Analysis (SVA) was used to point out the direction and width of improvement in order to increasing the efficiency. After the analysis, we had several conclusions. First, there were seven Overall Inefficiency medical nurse’s stations (46.67%) and six surgical nurse’s stations (46.15%). Second, there were four medical stations (26.67%) and 5 surgical stations (38.46%) Pure Technical Inefficiency. Third, there were seven medical nurse’s stations(46.67%) and six surgical nurse’s stations(46.15%) Scale Inefficiency.There were five and three Relative Inefficiency medical and surgical nurse’s stations and the rate of efficiency was between 0.9 and 1, which was Marginal Inefficiency Units. If we adjusted some input variants, they would become efficiency. There were five Relative Inefficiency stations’ efficiency rates were <.9, and were Distinctly Inefficiency Units. According to the Slack Variable Analysis, we focused on the relative inefficiency medical and surgery stations. The stations that need to decrease resource of the input variables included divisions of Hematology and Oncology, Neurology, Cardiopulmonary, Surgical Oncology, Orthopedic Surgery and Oral Maxillofacial Surgery. Besides Orthopedic Surgery and Oral Maxillofacial Surgery, after merging twice with the inefficiency of bed turnover rate, the other four stations had their specialized, more severe of patients, and more nurses depended which need more manpower of nurse.The surgical units need more nursing because preoperation and postoperative care. Therefore, the demand of nursing is more than others. To achieving the relative efficiency, the proper manpower should be arrange depending on the number of patients having surgery. Therefore, when consideration of Scale Efficiency at the same time, to achieve the overall efficiency, the adjustment of the input and output were not only decreasing the input, but depending on each unit to modify the input and output. Nursing is a professional work and nurses should continue to update their knowledge and innovation. Therefore, it is important for the nurses to join the on-the-job education. Due to nurses should shift duty on work, the education should be diversification, such as setting a Powecam in the station and recoding it to the DVD, or using the digital synchronized or non-synchronized self-learning, to develop their professional characters. Besides, due to adjust the non-department policy for patients’ administration, nurses should take turn to change stations for cross departments learning to make them more flexible for work. Therefore, not only evaluation the severity of patients, we should reorganize the work of professional nursing work and non-nursing work (ex. nurse’s assistant). Using mixed manpower as partial nurses and nurse’s assistants (Skill-Mixed Model) executes the nursing and increases the hours of nursing on patient and their satisfaction. Under this foundation, those nurse’s assistants should receive correct and professional skill training in respect of nursing.