背景與目的：近年來醫院為提升急診品質，紛紛進行多方面的品質監測及改善措施，以提高醫療服務品質。醫院急診室最常見的問題是病人於72小時內返診，此不僅可能延誤病人診斷與就醫時效性，亦增加社會醫療資源的浪費。因此，本研究目的為探討急診病人非計畫性返診之重要影響因素。
研究方法：研究以中部某醫學中心，於2010年6月至9月之急診病人於看診結束後，出院72小時內返診的301位病人及陪同返診家屬為主要收案對象。將所收集資料先行編碼及登錄建檔，並運用電腦統計軟體SPSS 12.0套裝軟體以進行描述性統計、卡方檢定分析病人特性、陪同返診家屬特性及醫療服務因素與非計畫性返診之相關性，最後以羅吉斯迴歸分析探討影響急診病人非計畫性返診之重要因素。
研究結果：本研究分析急診病人72小時內非計畫性返診中以小於18歲占最多105人(34.9%)，陪同家屬的關係中以父母且31至45歲族群為最多數占129人(42.9%)。在非計畫性返診與醫療服務因素之相關性分析結果顯示，看診醫師等級、來診月份、病人之檢傷級數與非計畫性返診均達顯著相關（P<0.01；P<0.001；P<0.01），在控制多變項下，以羅吉斯迴歸分析影響急診病人非計畫性返診之重要因素，結果發現來診月份及檢傷級數均達顯著性（P<0.01）。
結論與建議：本研究結果發現，看診醫師等級、來診月份及病人之檢傷級數與非計畫性返診有高度相關，因此，建議未來對急診病人之醫療服務，可考慮由資深且具急診專業能力的醫師進行診療，在暑假期間急診病人高峰期應有適當之醫療人力安排，並對重症級數之病人應更加強醫療照護，同時對急診病人宜有出院後續關懷與追蹤，使出院返家之病人及家屬放心，以降低病人返診的機率，減少醫療資源浪費且促進醫療品質。
Background and Purpose: In recent years, wide ranges of monitoring devices and measurements have been carried out in hospitals in hopes of advancing the quality of emergency care. One of the most common problems met by emergency departments is the Unplanned Return Emergency Department (URED) in 72 hours. The URED might delay diagnosis, decrease quality of care and increase the cost. The purpose of this study was conducted to investigate the factors that contribute to the problem.
Methods: The study was conducted patients with URED in 72 hours from June 2010 to September 2010 in a medical center in central Taiwan. Data were coded after collected and descriptive statistics were compiled and analysed by Chi-square test for the relation between URED, patient family characteristcs and services quality of hospital under the software of SPSS 12.0. Logistic regression analysis were utilized to explore the factors of URED.
Results: There were 105 people (34.9%) less than 18 year old. One hundrant and twenty nine people (42.9%) were acccompanied by parents aged between 31 to 45 year old. Factors concerning medical care quality such as proficiency of physician, calendar month of visit, and patient triage level all showed correlation with rate of URED（P<0.01, P<0.001 and P<0.01, respectively）. After controlling variables, logistic regression showed a significant effect to the URED were difference month and triage level (P<0.01).
Conclusions and Recommendations: Results showed that proficiency of physician, calendar month of visit, and patient triage level were highly associated with unplanned visits to ED. It is therefore reasonable to suggest future emergency care to be carried out by experienced physicians, increasing medical staff during the peak summer months, and providing extra care to patients with higher triage level. Furthermore, following up of discharged patients provides assurance to the patients and families, thereby reducing revisit rate. Resulting in reduced wasting of medical resources and improved medical care quality.
