多重抗藥性菌株資訊自動化監測與應用

Surveillance System for Smart Management of Multidrug-resistant Organisms

對於抗生素的廣泛使用，造成醫院內抗藥細菌的散布，已成為醫藥界之重大難題。世界衛生組織呼籲各國共同對抗抗藥性細菌的危機，並訂為2011年世界健康日的主題。資訊管理系統能有效地支持感染預防與控制、減少抗藥性細菌散布。本報告以一2,200床醫學中心運用實驗室抗藥性細菌檢驗結果資料庫建立地域性自動判斷與群突發監測機制，架設系統後抗藥性細菌偵測正確率由63%提升到100%（P＜.001）。此外，套用管制圖將資料的質性、量性進行視覺化呈現，以便於輕鬆、有效掌控單位的多重抗藥性菌株資料，並根據統計分布能有縝密的思考，做為推測群聚風險程度之重要參考。在此研究中結合管制圖C chart警示標準設定，與聚合式階層分群法（agglomerative hierarchical clustering），可對抗藥性菌株群突發有很好的相關性（area under ROC curve, AUC, 0.93）。本研究顯示藉由資訊系統輔助可以即時有效的偵測抗藥性菌株群突發，節省的時間讓感控人員能著力於教育、提醒醫囑開立與病人安置、實地稽核及協助單位品管活動等務實的業務，幾年下來深耕與努力，多重抗藥性菌株個案之即時（24小時內）開立隔離醫囑率由2008年25.4%、2013年進步到82.2%；多重抗藥性菌株個案報告日和隔離醫囑開立之平均相距日數由307.8小時進步到18.1小時。因此，資訊系統的導入，可協助感染控制監測作業的時效和精確度，輔助人員做正確及快速的決策，以提升醫療品質與病人就醫的安全性。

The emergence and spread of multidrug-resistant organisms (MDROs) have caused significant challenges for infection control personnel. To resolve this problem, the WHO has made antimicrobial resistance an organization-wide priority and the focus of the 2011 World Health Day. Information technology is expected to improve efficiency in automated surveillance and infection control (1-8). The present study developed a Web-based MDRO surveillance and outbreak detection information system from electronic microbiological data at a 2200-bed major teaching hospital in Taiwan. The accuracy of MDRO detection was 63.9%±26.4%by infection control personnel; the system was 100% (P < .001). The optimal UCL for MDRO outbreak detection was the upper 90% confidence interval (CI) using germ criterion with clustering (area under ROC curve [AUC], 0.93). This study demonstrates that an internet-based MDRO surveillance and outbreak detection information system provides useful information to facilitate the timely targeting of the correct unit by infection control personnel for appropriate intervention. The proportion of contact precautions among incident patients increased after implementation of the system (16.5% versus 82.2%, P = .001). Time lag of contact isolation (hours) improved after implementation of the system (307.8 versus 18.1, P = .001). In such a system, visualization methods are also important for clearly presenting the proximity of time and space and species of MDRO, and for facilitating data-driven decision-making. Implementing this system could, therefore, improve patient safety as well as the quality of medical care in a hospital.