院內血流感染監視輔助系統之開發與建置

Development of a Guideline based System for Bloodstream Infection Surveillance and Control

江政家(Jheng-Jia Jiang)

指導教授 ： 劉建財

臺北醫學大學/醫學科技學院/醫學資訊研究所/碩士(2011年)

https://doi.org/10.6831/TMU.2011.00177

摘要

院內血流感染的發生不僅提高醫療院所的治療成本，也危害到病人的生命安全是醫療品質不可忽視的重要環節，隨著日常的臨床作業逐漸資訊化，院內血流感染的作業需要收集並查閱各種資料後比對收案標準進而確認感染個案的發生，然而審查個案所需要的相關資料是分散於各醫療系統之間，為減少資料查詢所消耗的人工作業時間，監視輔助系統可協助收集各項資料直接提供感控人員查詢。最後感控人員收集資料後再依據院內血流感染的收案標準判斷是否有感染個案的發生。

本研究將建置院內血流感染監視輔助系統，系統包含三個模組：(1)自動化模組；(2)電子化規則判斷模組；(3)資料作業模組。自動化模組是將院內血流感染資料的收集設計成自動化收集，所需要收集的資料內容是透過分析院內感染的收案流程與感染個案資料就可得知，資料內容包括陽性檢驗報告、微生物的抗生素敏感性試驗報告、用藥資料、症狀資料、住院資料。規則判斷模組透過探討將院內血流感染的收案標準設計成電子化判斷規則演算法的步驟，設計電子化判斷規則，規則包含血流感染的疑似收案規則與其他的可能感染部位的排除規則，然而結合自動化收集資料模組與規則判斷模組可判斷收集後的相關資料進而產生判斷後的警示訊息。資料呈現模組則是提供介面讓感控人員查詢系統所收集的病人資料與歷程結果。

本研究以回溯性研究(retrospective study)分析2011年3月- 5月的住院病人資料將監視輔助系統所警示的疑似血流感染個案與感控人員審查的實際血流感染個案相比較，住院病人共6334位，感染個案數為54人，系統提示人數為73人，結果顯示系統的疑似個案提示效果具有71%的敏感度(sensitivity)與100%的特異度(specificity)。

研究結果顯示監視輔助系統所提示的疑似個案共有73筆的結果，以此73筆的結果資料查詢相關的陽性檢驗結果只需要確認473筆，與原本的陽性檢驗清單紀錄(3209筆)相較而言，可節省85%的查詢時間。

關鍵字

院內感染 ； 血流感染 ； 演算法 ； 自動化

並列摘要

Healthcare-Associated Infection increased the cost of hospital treatment. Patient's life is an important part of quality of medical care. Bloodstream Infection (BSI) is common of Healthcare-Associated Infection. Infection Control professionals (ICPs) collect data relevant to bloodstream infection, and justify true infection cases. Because electronic information is distributed on many Hospital information systems, data collection is time-consuming and costly. It is important to build a Bloodstream infection Monitoring system for collecting patient's electronic medical records. The system can automatically perform data collection and case findings, It can provide more electronic information based on the CDC’s guideline.

The bloodstream infection monitoring system includes three modules：(1) Automated data collection module is to collect data on hospital information systems, the data include laboratory reports, susceptibility tests, medication records, symptom records, and administrative data. (2) Electronic surveillance algorithm module is for creating electronic surveillance algorithms based on the existing manual surveillance practices methods. (3) Bloodstream infection data view module is to provide infection control professionals to check the system to collect data on patients and history results.

In this study, retrospective analysis to validate the system was performed. The subjects were the patients admitted to a regional hospital from March 1 to May 30, 2011. Sensitivity and specificity are used for comparison between suspected BSI cases generated by the system and the true BSI cases found by an infection control review team. The total patients involved in the study was 6334, and the number of true BSI cases was 54. The system generated 73 suspected cases. Thus, the sensitivity and specificity of the system were 71% and 100%, respectively. Moreover, the infection control review team has to investigate only 473 positive lab tests for those 73 patients. Compare to the original lab tests with positive results (3209) the investigation effort can be saved up to 85%.

並列關鍵字

HAI ； Healthcare-associated infection ； Bloodstream infection ； Algorithms ； Automated

參考文獻

行政院衛生署疾病管制局. (2009). 醫療照護相關感染監測定義

衛生署疾病管制局. (2009). 醫療照護相關感染監測定義

行政院衛生署疾病管制局. (2008). 傳染病統計暨監視年報.

Anderson, D. J., Kirkland, K. B., Kaye, K. S., Thacker Ii, P. A., Kanafani, Z. A., Auten, G., et al. (2007). Underresourced Hospital Infection Control and Prevention Programs: Penny Wise, Pound Foolish? Infection Control and Hospital Epidemiology, 28(7), 767-773.

Bouam, S., Girou, E., Brun-Buisson, C., Karadimas, H., & Lepage, E. (2003). An intranet-based automated system for the surveillance of nosocomial infections: Prospective validation compared with physicians' self-reports. Infection Control and Hospital Epidemiology, 24(1), 51-55.