Feasibility and Diagnostic Accuracy of Self-Pulse Screening for Atrial Fibrillation Detection in Community-Dwelling Adults: A Prospective Cohort Study

Abstract Background Self-pulse palpation is a low-cost method to detect atrial fibrillation (AF) in community dwelling older adults, but its diagnostic accuracy in community settings remains uncertain. Methods We conducted a prospective cohort study embedded within a randomized clinical trial across ten general practices (urban and rural) in the west of Ireland. Community-dwelling adults aged ≥55 years with a CHA₂DS₂-VASc score >2 and deemed suitable by their general practitioner for AF screening and anticoagulation were included. Participants were taught to detect an irregular pulse and instructed to perform a 2-minute self-pulse check three times daily for two weeks, recording results in a diary. The reference standard was AF detected by one-week external loop recorder (ELR) monitoring or by ECG after clinician-detected irregular pulse. Results Of 488 participants who completed ELR cardiac monitoring, 467 (95.7%) completed self-pulse screening. Mean age was 75 (SD 7.0) years, 68% were female, and mean CHA₂DS₂-VASc score was 3. Participants performed a median of 40 pulse checks (IQR 37–42); 48% completed all scheduled pulse checks. An irregular pulse was reported by 172 participants (37%), and was recurrent (>1 episode) in 131 (28%). Compared to ELR-detected AF, self-reported irregular pulse yielded a sensitivity of 39% (95% CI 22–56%), specificity 63% (95% CI 59–68%), positive predictive value (PPV) 7% (95% CI 3–11%), and negative predictive value (NPV) 94% (95% CI 91–96%). Diagnostic performance did not improve when restricted to recurrent irregular pulses. Conclusion Serial self-pulse screening is feasible and acceptable in older adults, but diagnostic accuracy is suboptimal. Though commonly recommended by a number of community organisations our findings do not support its use as a stand-alone screening method for atrial fibrillation in the general public.