BACKGROUND AND OBJECTIVE: Several investigations, most of them cross-sectional, have analyzed the association between smoking and depression, but there are not enough evidences to establish the direction of the association. Our objective was to prospectively assess the association between smoking habit and depression incidence in the SUN (Seguimiento Universidad de Navarra) cohort. SUBJECTS AND METHOD: Dynamic cohort study (recruitment permanently open) based on 8,556 graduate participants where depression incidence was ascertained during 6 years of follow-up. Depression was defined as the presence of a self-reported physician diagnosis and/or use of antidepressant medication in at least one of the 2-year, 4-year or 6-year follow-up questionnaires. The association between smoking and depression incidence was estimated through the calculation of hazard ratios (HR) and their 95% confidence intervals (CI) using Cox regression models (proportional hazards). RESULTS: The mean follow-up period was 47.4 months. One hundred and ninety participants initially free of depression reported a physician diagnosis of depression during follow-up. A significantly higher risk was found for smokers (HR = 1.39; 95% CI, 1.00-1.93) when they were compared to non smokers, whereas an inverse association was found for ex-smokers who had quit smoking more than 10 years ago (HR = 0.42; 95% CI, 0.19-0.94). CONCLUSIONS: This study supports, with a prospective design, an increased risk of depression associated to smoking.