Because the risk factors for constipation in the general population are poorly defined, we examined the association between age, body mass index, and selected lifestyle and dietary factors with the prevalence of constipation in the Nurses' Health Study.A total of 62,036 women, aged 36-61 yr and free of cancer, responded to mailed questionnaires in 1980 and 1982, which assessed bowel movement frequency, dietary, and lifestyle factors. Constipation was defined as two or fewer bowel movements weekly.A total of 3327 women (5.4%) were classified as having constipation. Age and body mass index were inversely associated with constipation (p for trend < 0.0001). In multivariate analysis, women who reported daily physical activity had a lower prevalence of constipation (prevalence ratio [PR] = 0.56, 95% CI = 0.44-0.70). Women in the highest quintile of dietary fiber intake (median intake 20 g/day) were less likely to experience constipation (PR = 0.64, 95% CI = 0.57-0.73) than women in the lowest quintile (median intake 7 g/day). The PR for constipation for women who both reported daily physical activity and were in the highest quintile of fiber intake was 0.32 (95% CI = 0.20-0.54) compared with those with physical activity less than once weekly and those who were in the lowest quintile of fiber intake. Higher frequency of aspirin intake was associated with an increasing prevalence of constipation (PR = 1.38, 95% CI = 1.17-1.62). Current smoking and alcohol use were inversely associated with constipation. Our data suggest that moderate physical activity and increasing fiber intake are associated with substantial reduction in the prevalence of constipation in women.