Background: Atrial fibrillation (AF) is underdiagnosed and especially undertreated in China. We aimed to investigate the prevalence of unknown and untreated AF in community residents (≥65 years old) and to determine whether an education intervention could improve oral anticoagulant (OAC) prescription.
Methods and findings: We performed a single-time point screening for AF with a handheld single-lead electrocardiography (ECG) in Chinese residents (≥65 years old) in 5 community health centers in Shanghai from April to September 2017. Disease education and advice on referral to specialist clinics for OAC treatment were provided to all patients with actionable AF (newly detected or undertreated known AF) at the time of screening, and education was reinforced at 1 month. Follow-up occurred at 12 months. In total, 4,531 participants were screened (response rate 94.7%, mean age 71.6 ± 6.3 years, 44% male). Overall AF prevalence was 4.0% (known AF 3.5% [n = 161], new AF 0.5% [n = 22]). The 183 patients with AF were older (p < 0.001), taller (p = 0.02), and more likely to be male (p = 0.01), and they had a higher prevalence of cardiovascular disease than those without AF (p < 0.001). In total, 85% (155/183) of patients were recommended for OAC treatment by the established guidelines (CHA2DS2-VASc ≥ 2 for men; ≥ 3 for women). OAC prescription rate for known AF was 20% (28/138), and actionable AF constituted 2.8% of all those screened. At the 12-month follow-up in 103 patients (81% complete), despite disease education and advice on specialist referral, only 17 attended specialist clinics, and 4 were prescribed OAC. Of those not attending specialist clinics, 71 chose instead to attend community health centers or secondary hospital clinics, with none prescribed OAC, and 15 had no review. Of the 17 patients with new AF and a class 1 recommendation for OAC, only 3 attended a specialist clinic, and none were prescribed OAC. Of the 28 AF patients taking OAC at baseline, OAC was no longer taken in 4. Ischemic stroke (n = 2) or death (n = 3) occurred in 5/126 (4%), with none receiving OAC. As screening was performed at a single time point, some paroxysmal AF cases may have been missed; thus, the rate of new AF may be underestimated.
Conclusions: We demonstrated a noticeable gap in AF detection and treatment in community-based elderly Chinese: actionable AF constituted a high proportion of those screened. Disease education and advice on specialist referral are insufficient to close the gap. Before more frequent or intensive screening for unknown AF could be recommended in China, greater efforts must be made to increase appropriate OAC therapy in known AF to prevent AF-related stroke.