Physical activity is inversely associated with depressive symptoms, and cigarette smoking is positively associated with depressive symptoms. Data from the first National Health and Nutrition Examination Survey (NHANES I) and the NHANES I Epidemiologic Follow-up study were analyzed to determine whether the relationship between physical activity and self-reported distress (depressive symptoms as measured by the Center for Epidemiologic Studies Depression Scale) was different for cigarette smokers and nonsmokers. Logistic regression was used to calculate odds ratios (adjusted for age, race, sex, education, alcohol use, and perceived health status) for depressive symptoms (> or = 16) associated with physical activity and smoking status among 2,054 respondents. At baseline, the odds ratio for depressive symptoms was about 2 times higher for moderately active smokers and nonsmokers, and 3 times higher for low active smokers and nonsmokers, compared with highly active nonsmokers. For 1,132 persons with a low number of depressive symptoms (< 16) at baseline, the incidence of depressive symptoms after 7-9 yr of follow-up was about 2 times higher for low/moderately active smokers and nonsmokers than for highly active nonsmokers. The association between physical activity and the prevalence and incidence of depressive symptoms is not significantly modified by smoking status.