Usability of an Atrial Fibrillation Anticoagulation Decision-Support Tool

• Main Authors: Mark L. Wess MD, SM, Jason J. Saleem PhD, MS, Joel Tsevat MD, MPH, Sara E. Luckhaupt MD, MPH, Joseph A. Johnston MD, MSc, Ruth E. Wise MSN, Jonathan E. Kopke, Mark H. Eckman MD, MS
• Format: Article
• Language: English
• Published: SAGE Publishing 2011-04-01
• Series: Journal of Primary Care & Community Health
• Online Access: https://doi.org/10.1177/2150131910387608

Introduction : In individuals with nonvalvular atrial fibrillation, anticoagulant therapy with warfarin reduces the rate of thromboembolic events but increases the risk of bleeding. Treatment decisions frequently are inconsistent with guidelines. A new web-based atrial fibrillation decision-support tool (AF-DST) provides patient-specific information on the risk-benefit tradeoff of anticoagulation. Methods : The authors performed a pilot usability testing study of the AF-DST with 4 medical house officers and 4 attending physicians by simulating 9 outpatient clinical encounters involving tradeoffs between risks and benefits of anticoagulation. They recorded positive and negative critical incidents in the simulations and assessed satisfaction with use of the AF-DST by the Computer System Usability Questionnaire (CSUQ; score range on each item: 1 = strongly disagree to 7 = strongly agree ). Results : Users found the AF-DST to be helpful and had high CSUQ scores (mean item score, 6.3). Usability testing identified 6 positive and 14 negative critical incidents. Participants felt that the AF-DST guided them toward the correct decision. Nevertheless, they desired more information on the “black box” calculations and ignored alerts. Training level appeared to affect how the AF-DST was used, in particular, how users interacted with the AF-DST. Conclusions : Overall satisfaction with the AF-DST was high and the tool effectively communicated recommendations and uncertainty. Usability testing identified design issues and potential errors caused by decision-support tool use; these gaps should be addressed prior to clinical implementation.