利用申報與行政資料進行醫療照護相關感染個案判定

dentifying Healthcare-Associated Infections Identification by Claims and Administrative Database

醫療照護相關感染一直是醫界相當重視的議題，學界與醫界也不斷進行相關研究，希望可以減低感染的發生。傳統上，大規模的健康照護相關感染研究，都藉由行政資料或申報資料來進行，而資料庫中的國際疾病分類碼，更是經常用來作為感染個案判定依據。然而許多研究證實，行政資料或申報資料中的國際疾病分類碼，常會受到內外在因素的影響，因而造成編碼錯誤。因此近年來有研究開始利用行政資料庫來發展感染個案判別模式，然而不論是在臺灣或是國際上，都缺乏相關文獻之整理。因此，本研究透過Medline文獻資料庫，搜尋2000年以後至2012年有關發展感染個案判別模式之研究。經過逐篇審視，共找到26篇相關研究。依據其發展類型，又可分為三種判別模式。本研究針對此三種判別模式進行介紹與比較，並對臺灣後續研究提出相關建議。

Healthcare-associated Infections was an important issue for medical society, many studies were conducted in academic and medical societies, to try to reduce the prevalence of healthcare-associated Infections. Traditionally, large-scaled studies were conducted by administrative or claims database, the infection cases were usually identified by the International Classification of Diseases system. However, many studies indicated that the coding bias of International Classification of Diseases codes might be caused by internal and external factors. In recent years, some researchers devoted themselves to infection cases identification model development in administrative database. However, the review of this topic was insufficient in literature. In this study, we retrieved studies from Medline databases from 2000 to 2012 and completed a manual review of cited articles to identify studies related to infection cases identification model development. A total of 26 studies were included in this review. In summary, there are three types of infection cases identification models development. We introduced these three types of models and provided some reflections and new ideas for future researches in Taiwan.