OBJECTIVE: We analyzed the cross-sectional association between obesity and smoking habits, taking into account diet, physical activity, and educational level. RESEARCH METHODS AND PROCEDURES: We used data from the 2002 Swiss Health Survey, a population-based cross-sectional telephone survey assessing health and self-reported health behaviors. Reported smoking habits, height, and weight were available for 17,562 subjects (7844 men and 9718 women) >/=25 years of age. BMI was calculated as (self-reported) weight divided by height(2). RESULTS: Mean BMI was 25.1 kg/m(2) for non-smokers, 26.1 kg/m(2) for ex-smokers, 24.6 kg/m(2) for light smokers (1 to 9 cigarettes/d), 24.8 kg/m(2) for moderate smokers (10 to 19 cigarettes/d), and 25.3 kg/m(2) for heavy smokers (>/=20 cigarettes/d) in men and 24.0, 24.1, 22.9, 22.9, and 23.3 kg/m(2), respectively, in women. Obesity (BMI >/= 30 kg/m(2)) was increasingly frequent with older age, lower physical activity, lower fruits/vegetables intake, and lower educational level. Compared with non-smokers, the odds ratio for obesity vs. normal weight (BMI = 18.5 to 25.0 kg/m(2)) adjusted for age, nationality, educational level, leisure time physical activity, and fruit/vegetable intake were 1.9 (95% confidence interval: 1.5 to 2.3) for ex-smokers, 0.5 (0.3 to 0.8) for light smokers, 0.7 (0.4 to 1.0) for moderate smokers, and 1.3 (1.0 to 1.7) for heavy smokers in men and 1.3 (1.1 to 1.6), 0.7 (0.5 to 1.0), 0.8 (0.5 to 1.0), and 1.1 (0.8 to 1.4), respectively, in women. DISCUSSION: Among smokers, obesity was associated in a graded manner with the number of cigarettes daily smoked, particularly in men. More emphasis should be put on the risk of obesity among smokers.