Many studies have demonstrated an association of both a sedentary lifestyle and a high body mass index (BMI) with greater risk for cardiovascular disease. Within the prospective SAPALDIA cohort (Swiss cohort Study on Air Pollution and Lung Diseases in Adults), we investigated whether regular exercise was protective against reduced heart rate variability (HRV), a clinically relevant predictor of cardiovascular morbidity and mortality, and whether adverse effects of obesity and weight gain on HRV were modified by regular exercise. Twenty-four-hour electrocardiograms were recorded in 1,712 randomly selected SAPALDIA participants aged >or=50, for whom BMI was assessed in the years 1991 and 2001-2003. Other examinations included an interview investigating health status (especially respiratory and cardiovascular health and health relevant behaviours including physical activity) and measurements of blood pressure, body height and weight. The association between regular physical activity and HRV and interactions with BMI and BMI change was assessed in multivariable linear regression analyses. Compared to sedentary obese subjects, SDNN (standard deviation of all RR intervals) was 14% (95% CI: 8-20%) higher in sedentary normal weight subjects; 19% (CI: 12-27%) higher in normal weight subjects exercising regularly >or=2 h/week; and 19% (CI: 11-28%) higher in obese subjects exercising regularly >or=2 h/week. Compared with sedentary subjects who gained weight, those who gained weight but did exercise regularly had a 13% higher SDNN (CI: 7-20%). Regular physical exercise has strong beneficial effects on cardiac autonomic nervous function and thus appears to offset the negative effect of obesity on HRV.