Abstract 9886: Use of a Smartwatch and App Designed by Stroke Survivors for Atrial Fibrillation Detection in Older Adults After Stroke/Transient Ischemic Event: Preliminary Findings from an Ongoing Randomized Clinical Trial

Background: Atrial fibrillation (AF) is a common cause of ischemic stroke that, though critical to diagnose and treat, often goes undetected. Smartwatches are capable of AF detection, but little is known about their use among stroke survivors. In collaboration with stroke survivors and clinicians, we designed a smartphone-smartwatch system for AF detection. In our ongoing clinical trial, we are examining the performance and user experience of the system in older stroke survivors. Hypothesis: The system will be accurate for AF detection among stroke survivors compared to 14-day ECG patch monitoring; participants will find the system usable and will adhere to daily use over 30 days. Methods: The Pulsewatch study is an ongoing two-phased randomized controlled trial with randomization at two timepoints into control or intervention groups. Phase I (14 days) assesses accuracy and usability of the smartwatch and study app. Phase II (30 days) examines watch wear adherence. The intervention group receives a smartwatch with embedded AF detection and a smartphone app for rhythm and symptom tracking. Eligibility criteria are: > 50 years old, ischemic stroke in the past decade, and no major contraindications to anticoagulation therapy. System accuracy for AF detection is compared to an ECG patch monitor, and usability is assessed via the System Usability Scale. Adherence is defined as daily watch wear time over the monitoring period. Results: To date, 69 participants have been randomized into phase I and 36 into phase II (mean age 65, 41% female, 87% White). The smartwatch system demonstrated a sensitivity of 66.7% (95% CI: 9.4 - 99.2) and specificity of 92.6% (95% CI: 82.1 - 97.9) for detection of 3 AF cases over the 14 day period (91.2% accuracy) compared to the ECG patch. The mean usability score was 65/100 and on average, participants wore the watch on 12/30 days for an average of 1.8 hours (SD 0.7) daily. Conclusion: A smartwatch and app dyad demonstrated reasonable accuracy for AF detection and high usability, though adherence was modest. Use of commercial wearables for AF detection among stroke patients shows promise, but additional data from the ongoing Pulsewatch clinical trial will continue to inform strategies to maximize adherence to these devices among older adults.