Encoding and Verification of a Computer-Interpretable Guideline: A Case Study of Pressure-Ulcer Management. Issue 1 (March 2015)

• Record Type: Journal Article
• Authors: Kim, Hyun-Young; Park, Hyeoun-Ae; Cho, InSook; Kim, JeongAh; Lee, JaeHoon
• Is Part Of: Health information management journal. Volume 44:Issue 1(2015)
• Page Start: 39
• Page End: 48
• Publication Date: 2015-03
• Subjects: Clinical Decision Support Systems -- Practice Guidelines as Topic -- Pressure Ulcer -- Patient Care Management -- Clinical Practice Guidelines -- Decision Support Techniques -- Knowledge Representation (Computer) -- Software Verification; Medical records -- Management -- Periodicals; Medical records -- Management -- Australia -- Periodicals; Medical records -- Management; Australia; Periodicals; Periodicals; 651.50426105
• DOI: 10.1177/183335831504400105
• Languages: English
• ISSNs: 1833-3583

Abstract

This study examined ways to improve the accuracy of translating clinical practice guidelines (CPGs) into a computer-interpretable guideline (CIG) for pressure-ulcer management using the Shareable Active Guideline Environment (SAGE) guideline model, and aimed to verify the accuracy of the obtained CIG. The study was conducted using the following procedures: selecting CPGs, extracting rules from the selected CPGs, developing a CIG using the SAGE guideline model, and verifying the obtained CIG with test cases using an execution engine. The CIG for pressure-ulcer management was developed based on 38 rules and three algorithms at the semiformal representation level using MS Excel and MS Visio. The CIG was encoded by two Activity Graphs consisting of 115 instances representing algorithms and rules as knowledge elements in the SAGE guideline model. Two errors were found and corrected. Results of the study demonstrated that a CIG representing knowledge on pressure-ulcer management can be effectively developed using commonly available programs and the SAGE guideline model, and that the obtained CIG can be verified with a locally developed execution engine. The CIG developed in the study could contribute to health information management once it is implemented successfully in a clinical decision support system.