Consumer Wearable Devices for Atrial Fibrillation Detection: A Systematic Review and Meta-Analysis of Diagnostic Accuracy

Abstract

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, affecting approximately one in five individuals after the age of 45, and is associated with a fivefold increase in stroke risk. A significant proportion of AF cases remain undiagnosed due to the paroxysmal and asymptomatic nature of the condition, and conventional diagnostic methods such as 12-lead ECG and Holter monitoring are limited in their ability to capture intermittent episodes. Consumer wearable devices, including smartwatches capable of photoplethysmography (PPG) and single-lead electrocardiography (ECG), have emerged as a potential tool for population-level AF screening. We conducted a systematic review and meta-analysis of 45 studies encompassing 1,281,855 patients to evaluate the diagnostic accuracy of consumer wearables for AF detection. Pooled sensitivity was 0.92 (95% CI 0.82–0.96) and pooled specificity was 0.98 (95% CI 0.97–0.99). Sample-size-weighted averages showed comparable accuracy for PPG-based and ECG-based devices: sensitivity 91.7% (95% CI 89.3-93.6%) vs 95.1% (95% CI 93.8-96.1%), and specificity 97.9% (95% CI 96.9-98.5%) vs 95.3% (95% CI 94.6-95.9%). These findings suggest that PPG and single-lead ECG function as complementary rather than competing modalities, with PPG suited to passive, continuous monitoring and single-lead ECG to on-demand rhythm confirmation. A combination of both technologies represents the most reliable approach to consumer wearable-based AF screening, though PPG-based alerts should be treated as a prompt for confirmatory testing rather than a standalone diagnosis.