Evaluating Arrhythmia Alerts from Smartwatches in Asymptomatic Emergency Patients

Smartwatches equipped with photoplethysmography (PPG) sensors are increasingly used for arrhythmia detection, yet their diagnostic accuracy in asymptomatic patients presenting to the emergency department (ED) remains underexplored. This study aimed to evaluate the performance of PPG-equipped smartwatches in detecting arrhythmias among asymptomatic patients and compare smartwatch-generated alerts with standard 12-lead electrocardiograms (ECG). This retrospective observational study included 523 asymptomatic patients who presented to a tertiary care ED over a one-year period with smartwatch-generated arrhythmia alerts. All patients underwent a standard 12-lead ECG upon arrival. Arrhythmias were categorized as atrial fibrillation (AF), sinus bradycardia, bundle branch block (BBB), or ischemic changes. Diagnostic metrics including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall accuracy were calculated for each condition. Misclassification rates were analyzed to assess smartwatch limitations. Of the patients, 35.27% had abnormalities detected by ECG, including AF (18.2%), sinus bradycardia (10.9%), BBB (17.5%), and ischemic changes (25.5%). For AF, smartwatches demonstrated a sensitivity of 53.85%, specificity of 69.06%, PPV of 28.87%, and NPV of 86.52%, with an overall accuracy of 66.18%. Detection rates were lower for sinus bradycardia (sensitivity 30.36%) and ischemic changes (sensitivity 4.88%). Misclassification rates were particularly high for ischemic changes (95.7%), sinus bradycardia (70%), and BBB (75%). While smartwatches exhibit moderate utility for AF detection, their limited accuracy for other arrhythmias underscores the need for algorithmic improvements. Despite these limitations, smartwatches may serve as supplementary tools to encourage early medical attention in asymptomatic patients.