利用更改藥名標示來降低調劑錯誤率之成效評估

Outcomes of Changing Medication Labeling to Reduce Dispensing Errors

如何建置一個防錯的醫療作業安全系統，愈來愈受重視。藥師除了執行處方適當性評估與用藥安全把關之外，對於藥師調劑與給藥的正確性，也希望有完善規劃，避免因作業疏失影響病患療效。藉由部內調劑疏失的通報紀錄，分析藥師調劑錯誤的原因，並研擬改善方法，降低調劑錯誤的發生，以確保病人用藥安全。針對藥師在處方調劑過程中，造成錯誤的原因加以探究，進而尋求改善之道。第一階段資料收集自2008年1月至10月分析原因並研擬改善方法，第二階段自2009年1月至10月，全面更改藥名後，並分析改善結果。統計藥名更改前後門診藥局、急診藥局及住院藥局調劑總筆數：3,575,135筆vs 4,500,659筆，疏失通報總筆數共2448筆vs 4839筆。疏失通報發生率：幾近疏失發生率為0.063％ vs 0.1％；調劑錯誤率為0.005％ vs 0.0023％ （p<0.001）。通報疏失原因中比較更改藥袋藥名前後錯誤型態，品名雷同18.4％ vs 17％ （p=0.25）、同成分不同劑量規格14％ vs 6.5 ％ （p<0.001）、同成分不同劑型7.8％ vs 3.5％ （p<0.001）、藥品（包裝）外形相像5.7％ vs 5.8％ （p=0.86）及位置相近15％ vs 12％ （p=0.001）等。重整藥名標示後，雖由學名改以商品名為主，在品名雷同及藥品（包裝）外形相像方面無顯著成效，因有些商品名也雷同；至於藥品（包裝）外形，因大部份同一藥廠之藥品在外觀及包裝上相似度高，僅能藉由宣導及相關提醒，或製作易混淆藥品標示提醒易疏失之品項。但經由更改藥名標示對劑量規格及同成分不同劑型的提醒可看到疏失率有意義之降低。另外執行雙人覆核的制度是非常重要的，研究顯示＞98％的調劑疏失可經由最後的覆核攔截下來。

Many dispensing errors occur in the hospital pharmacy and these can endanger patients. Therefore, setting prevention approaches for dispensing error is urgent. To examine the type and reasons for dispensing errors collected via intranet spontaneous reporting system and to evaluate the efforts implemented to reduce overall medication-related errors. Modifications of the medication labeling on dispensing sheet and package were made: 1. Trade names were used as the identification marker. 2. Tall man letters were applied in a variety of visual presentations of drug names. 3. Special instructions (e.g. high, intermediate and low) were placed directly behind the medication name. Dispensing errors collected from the pre- and post-change periods of Jan to Oct. 2008 and Jan. to Oct 2009, respectively, were analyzed. The total dispensed medications during the pre- and post-change periods were 3,575,135 and 4,500,659 respectively. The number of errors reported for the pre- and post-change periods were 2448 (0.068%) and 4839 (0.11%). Of these incidents, dispensing errors occurred at a rate of 0.005% and 0.0023%, and near miss occurred at a rate of 0.063% and 0.1%, respectively, for the pre- and post-change periods. The incidence of errors associated with similarly package drug labels was similar (5.7% vs.5.8%), however, differences were found in errors associated with look-alike medication name (18.4% vs.17%; p=0.25), similar name with different dosages (14% vs. 6.5%; p<0.001) and similar name with different dosage forms (7.8% vs. 3.5%; p<0.001) for the two periods. It was also found that up to 98% of dispensing errors can be intercepted if a double check is conducted. Proactive assessment of potential for medication errors can reduce the frequency and consequences of errors. Progress is being made on preventing confusing names, labels, and packages from making it to pharmacy shelves and hospital wards in the first place, and computerized order entry and bar coding system should help to decrease dispensing errors in the future.