Background: Cerebrovascular disease is associated with depression in later life. Smoking is a known risk factor for cerebrovascular disease and, as a consequence, may contribute to the development of depression in the elderly. This study was designed to investigate the association between smoking and depression in people aged 60 years or over.Methods: Cross-sectional survey of older adults attending a representative sample of general practitioners in Western Australia. Subjects were divided into four groups according to their smoking status: never smoked, ex-light smoker, ex-heavy smoker (> 20 cigarettes/day), and current smoker. Depressive symptoms were assessed with the Center for Epidemiologic Studies-Depression Scale (CES-D). CES-D score of 22 or more was used to define the presence of clinically significant depression.Results: 1030 completed the assessment. Their age ranged from 60 to 101 years and 57.2% were female. The mean +/- S.D. CESD score was 10.5 +/- 8.1, 10.6 +/- 8.3, 12.5 +/- 10.2 and 13.1 +/- 11.0 for never smokers, ex-light smokers, ex-heavy smokers and current smokers respectively (p = 0.037), with 7.7%, 8.5%, 13.8% and 17.2% having CES-D >= 22 (p = 0.016). Current or past heavy smoking was associated with increased risk of clinically significant depression when compared to never or past light smoking (OR = 1.58, 95% CI = 1.01-2.48-after adjustments were made for age, gender, place of birth, social isolation, self-perceived health and harmful or hazardous drinking).Conclusion: Past heavy smoking and current smoking are associated with increased frequency and severity of depression. Smoking cessation may play an important role in reducing the burden of depression in later life, but the success of smoking cessation interventions in decreasing the incidence and prevalence of depression might be predicated on the timing of the intervention; i.e., before the psychobiological changes associated with smoking become irreversible. (c) 2005 Elsevier B.V All rights reserved.