There is increasing evidence that hypertension may contribute to the development of dementia. We investigated the relation of antihypertensive drug use and the risk of dementia in the cohort of the population based Rotterdam Study. The study cohort included 7046 elderly, free of dementia at baseline. Dementia was diagnosed in a stepwise procedure. Participants were first screened. Screen positives were further tested. Those suspected of dementia underwent a diagnostic work-up. Dementia and its subtypes were diagnosed according to prevailing criteria. A Cox proportional hazards model was used to estimate relative risks. After a mean follow-up of 2.2 years, subjects taking antihypertensive medication at baseline (n = 2015) had a reduced incidence of dementia (adjusted relative risk, 0.76; 95% confidence interval 0.52-1.12). This risk reduction was most pronounced for vascular dementia, (adjusted relative risk, 0.30; 95% confidence interval 0.11-0.99). For Alzheimer's disease the relative risk was 0.87, not significant. Dementia may be prevented by antihypertensive treatment. In order to confirm any relation in Alzheimer's disease larger observational studies with longer follow-up are needed.