Diagnostic accuracy of wearable technologies for atrial fibrillation detection: a systematic review and meta-analysis

Abstract

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia and a major cause of ischemic stroke. Because AF is frequently paroxysmal and asymptomatic, conventional short-term monitoring often fails to detect it. Wearable technologies have emerged as promising tools for AF screening, but their diagnostic performance requires comprehensive evaluation. We conducted a systematic review and diagnostic test accuracy meta-analysis following PRISMA-DTA guidelines. PubMed, Embase, and the Cochrane Library were searched from inception to 25 February 2026 for prospective and retrospective studies using electrocardiographic reference standards. Bivariate random-effects and Bayesian hierarchical models were used to pool diagnostic accuracy estimates. Fifty-one studies involving 33,374 participants from 15 countries were included. PPG-based smartwatches demonstrated the highest sensitivity (96.3%) with specificity of 93.6%. ECG-based smartwatches achieved sensitivity of 89.0% and specificity of 92.8% (DOR 103.9). Smartphone-based ECG devices showed the strongest overall diagnostic performance, with sensitivity of 94.5%, specificity of 95.8%, DOR of 396.4, LR+ of 22.6, and LR− of 0.057. Smartphone-based PPG applications achieved sensitivity of 92.0% and specificity of 91.9%. Apple Watch demonstrated sensitivity of 94.1% and specificity of 94.6%. Meta-regression revealed no statistically significant differences in diagnostic accuracy between device categories despite slightly higher point estimates for PPG-based smartwatches. No significant publication bias was detected (Deeks' test, p=0.18). Wearable devices demonstrate excellent diagnostic accuracy for AF detection, particularly smartphone-based ECG applications. These technologies support scalable AF screening and monitoring, although heterogeneity in study populations and diagnostic thresholds should be considered when applying these findings to routine clinical practice.