Individual and combined impact of lifestyle factors on atrial fibrillation in apparently healthy men and women: The EPIC-Norfolk prospective population study

Background Lifestyle factors are important targets for prevention. The cumulative impact of healthy lifestyle on atrial fibrillation in the population has not been quantified. Design Prospective population-based cohort study. Methods Four lifestyle factors (normal weight, currently not smoking, no or moderate alcohol intake, and physically not inactive) were assessed in apparently healthy 21,499 men and women aged 39-79 years participating in the EPIC study in Norfolk, UK. The age and sex-adjusted hazard (95% confidence interval) of hospital admission with a diagnosis of atrial fibrillation during an average follow-up of 17.1 years was examined for each factor separately and for a health score comprising factors with significant impact. Results Normal weight, currently not smoking and low alcohol intake were associated with a significantly lower risk of atrial fibrillation, whereas not being physically inactive showed no significant association. We used a score of one point each for not smoking, low alcohol intake and body mass index 25 to 27.5kg/m(2), and two points for body mass index<25kg/m(2). Compared with men and women with four health points, hazard ratios of atrial fibrillation were 1.25 (1.11-1.41), 1.56 (1.39-1.75), 1.83 (1.56-2.16) and 2.82 (1.85-4.29) for participants with three, two, one and no health points, respectively (p<0.0001 for trend). Results were consistent by sex, age, education level, social class and after excluding participants with hypertension and diabetes. Conclusion Three lifestyle factors combined predict an almost 2.8-fold difference in the risk of atrial fibrillation in men and women.