Aims: To elucidate the relationship between tobacco smoking and depression, and to estimate the impact of other substance dependencies. Design: Cross-sectional. Participants: A total of 1,849 men and women were interviewed face-to-face using a validated structured questionnaire. According to their tobacco smoking behavior, participants were grouped into never smokers, ex-smokers and current smokers. Measurements: Data were generated through the WHO/ISBRA study, an international multicenter study with a cross-sectional design. A standardized questionnaire was administered face-to-face by trained interviewers. Logistic regression analysis was used to predict depression. Results: There was a significant difference across the 3 smoking groups in the number of subjects who had major depression (DSM-IV) during their lifetime. The highest rate of depressives was found in current smokers (23.7%), the lowest rate in never smokers (6.2%), while the rate of those who had quit smoking (14.6%) was between both. In a logistic regression analysis, alcohol dependence (both current and during lifetime) as well as cocaine dependence were significant predictors of depression. However, the association between smoking and depression still remained statistically significant. Conclusions: This study adds support to the evidence that smoking is linked to depression. It also elucidates the importance of taking into account alcohol and cocaine dependence since they have a significant impact on the relationship between smoking and depression.