Post-Fielding Surveillance of a Guideline-Based Decision Support System

摘要

Quality assurance (QA) processes for new technologies are used to ensure safety. Clinical decision support systems (DSS), identified by the Institute of Medicine (IOM) as an important tool in preventing patient errors, should undergo similar predeployment testing to prevent introduction of new errors. Post-fielding surveillance, akin to post-marketing surveillance for adverse events, may detect rarely occurring problems that appear only in widespread use. To assess the quality of a guideline-based DSS for hypertension, ATHENA DSS, researchers monitored real-time clinician feedback during point-of- care use of the system. Comments (n=835) were submitted by 44 of the 91 (48.4 percent) study clinicians (median 8.5 comments/clinician). Twenty-three (2.8 percent) comments identified important, rarely occurring problems. Timely analysis of such feedback revealed omissions of medications, diagnoses, and adverse drug reactions due to rare events in data extraction and conversion from the electronic health record. Analysis of clinician-user feedback facilitated rapid detection and correction of such errors. Based on this experience, new technologies for improving patient safety should include mechanisms for post-fielding QA testing.