利用醫療資訊技術建立醫囑藥品交互作用提示系統之價值，已廣受各醫療院所肯定。藥品交互作用提示系統不僅具專業知識教育功能外，亦對病患用藥安全提供多一層保障。本研究之目的在了解某醫學中心門診醫囑藥品交互作用系統之成效，並探討醫師行為之改變，做為系統提昇之參考。
本研究利用交互作用提示系統之回饋管理輔助功能，收集某醫學中心2002年1月至2004年3月，門診處方筆數及醫囑藥品交互作用提示系統所得之資料，進行系統提示率及使用者接受率之分析。分析系統提示率及使用者接受率各月之間的差異，以及系統提示率與醫師接受率之間的關聯性。並進一步分析使用者不接受之藥品交互作用提示項目分佈情形，並以問卷調查使用者的意見。
醫囑藥品交互作用提示系統平均提示率為0.228 % 0.117 %，使用者平均接受率為4.737 % 0.542 %，各月之間無明顯差異（p<0.01）。系統提示率愈高使用者接受率愈低，相關係數為-0.518（p<0.01）。醫師不接受之藥品交互作用提示項目中，毛地黃配醣體（digitalis glycoside）與thiazide類利尿劑之交互作用提示項目發生比例佔46.54%。醫師接受之藥品交互作用提示項目中，quinolones與制酸劑（antacids）佔30.87%，tetracyclines與制酸劑（antacids）佔11.84%。問卷調查結果，認為不需再提示為3.43%，繼續提示為68.83%，由系統強制不可併用為8.92%，沒有意見為16.98%。雖然藥品交互作用即時提示系統使用者之接受率低，但多數使用者仍希望系統繼續提供藥品交互作用訊息。
然而系統的精確度與互動性會影響醫師接受率，因此，系統之互動式應更高：如加入疾病診斷碼、臨床檢驗數值等作為判斷之依據，並且加入視覺化系統模式以避免因提示疲乏而造成接受度之降低。而針對含有潛在性交互作用之處方，藥師應加強監控病人臨床反應，並給予病患適當的藥物用藥指導，以確保病人用藥安全。
It has been wildly approved by hospitals to build the drug-drug interaction warning systems with medical informatics. The systems provide not only professional education, but also more safety for patients’ drug use. This research aims at the efficacy of the drug-drug interaction warning systems, as well as the users’ behavior change in out-patients clinics at a general teaching hospital. The results of this research will help to upgrade the systems.
By using part of feedback management support in the drug-drug interaction warning system, this research collected a number of out-patients orders and data from a general teaching hospital from January 2002 to April 2004. We tried to analyze the system-warning rates, the user-accepted rates, the monthly correlations between the system-warning rates and the physician-accepted rates. Furthermore, we analyzed the items rejected by the users, and required their opinions by questionnaires.
The mean warning rate is 0.228% 0.117% and the mean acceptation rate is 4.737% 0.542%. There is not significant difference between each month (p<0.01). The system-warning rate is negative correlation with the user-accepted rate. The coefficient score is –0.518 (p<0.01). Based on the rejected items, the rate of digitalis glycoside-thiazide is 46.54%, the quinolone-antacids is 30.87%, and the tetracyclines-antacids is 11.84%. As the result of the questionnaires, 3.43% users think the systems are useful, nearly 9% users suppose the orders will be rejected as soon as the drug-drug interaction warning systems are launched, and 16.98% users have no comments.
Although the acceptation rate of the warning systems is low, most of the users still hope the systems can continuously provide the information of drug-drug interaction. However, the accuracy of the systems and the interactivity affect the preference of the users, so the systems need further improvements, such as making decisions with ICD 9 code or the laboratory data, and adding the visualization technologies to avoid the exhaustion arisen by over-warming. As far as the orders with potential interactions, pharmacists should closely monitor the patients’ clinical reactions, and offer them proper medical directions in order to secure the correct drug usages.
