Background: The relative and absolute risks of stroke and heart failure attributable to atrial fibrillation (AF) have not been sufficiently examined.

Methods: A prospective study of 23,731 community-dwelling Japanese individuals was conducted. Participants were divided into two groups based on the presence or absence of prevalent AF (n = 338 and n = 23,393, respectively). Excess events (EE) due to AF and relative risks (RRs) determined using the non-AF group as the reference for incident stroke and heart failure were estimated using Poisson regression stratified by age groups (middle-aged: 40–69 years old; elderly: 70 years of age or older) after adjustment for sex and age.

Results: There were 611 cases of stroke and 98 cases of heart failure during the observation period (131,088 person-years). AF contributed to a higher risk of stroke both in middle-aged individuals (EE 10.4 per 1000 person-years; RR 4.88; 95% confidence interval [CI], 2.88–8.29) and elderly individuals (EE 18.3 per 1000 person-years; RR 3.05; 95% CI, 2.05–4.54). AF also contributed to a higher risk of heart failure in middle-aged individuals (EE 3.7 per 1000 person-years; RR 8.18; 95% CI, 2.41–27.8) and elderly individuals (EE 15.4 per 1000 person-years; RR 7.82; 95% CI, 4.11–14.9). Results obtained from multivariate-adjusted analysis were similar (stroke: EE 8.9 per 1000 person-years; RR 4.40; 95% CI, 2.57–7.55 in middle-aged and EE 17.4 per 1000 person-years; RR 2.97; 95% CI, 1.99–4.43 in elderly individuals; heart failure: EE 3.1 per 1000 person-years; RR 7.22; 95% CI, 2.06–25.3 in middle-aged and EE 14.1 per 1000 person-years; RR 7.41; 95% CI, 3.86–14.2 in elderly individuals).

Conclusions: AF increased the risk of stroke by the same magnitude as that reported previously in Western countries. AF increased the RR of heart failure more than that in Western populations.