We studied the hypothesis that the BclI polymorphism of the glucocorticoid receptor gene is associated with ail increased probability of being a (heavy) smoker and a decreased ability to quit smoking. The study cohort consisted of all subjects in the Rotterdam Study, a Dutch population-based cohort of people aged 55 years and older, for whom BclI genotyping and smoking status at baseline were available. In prospective analyses, the smoking status was reassessed during three additional examination rounds. Logistic regression analysis was used to study the association between BclI polymorphism and being a smoker or a heavy smoker at baseline. Furthermore, the relationship between BclI polymorphism and incident smoking cessation was tested with Cox proportional hazards analysis within those who smoked at baseline. In total, 6358 subjects were included in the study. The presence of a G-allele was not associated with current smoking at baseline [odds ratio (OR) = 0.96, 95%confidence interval (CI): 0.85-1.09] or with the incidence of smoking cessation during follow-up [hazard ratio (HR) = 0.98, 95%CI: 0.80-1.19]. Within Current smokers. having a G-allele was not significantly associated with the risk of being a heavy smoker when measured by pack-years smoked (OR = 1.07, 95%CI: 0.85-1.35) or daily consumption of tobacco (OR = 1.10, 95%CI: 0.88-1.37). We were not able to replicate the earlier findings indicating that the proportion of current smokers is lower among carriers of the CC-genotype of the BclI glucocorticoid receptor. Furthermore. the BclI glucocorticoid receptor polymorphism did not predict the incidence of smoking cessation in the general elderly population.