There are a few studies that report cognitive impairment as a complication of treatment with beta-blockers. We aimed to evaluate the longitudinal association between use of beta-blockers, as a class, and incident risk of all-cause dementia, vascular dementia, Alzheimers and mixed dementia in the prospective population-based Malmo Preventive Project. We included 18,063 individuals (mean age 68.2, males 63.4%) followed up for 84,506 person-years. Dementia cases were retrieved from the Swedish National Patient Register and validated by review of medical records and neuroimaging data. We performed propensity score matching analysis, resulting in 3720 matched pairs of beta-blocker users and non-users at baseline, and multivariable Cox proportional-hazards regression. Overall, 122 study participants (1.6%) were diagnosed with dementia during the follow-up. Beta-blocker therapy was independently associated with increased risk of developing vascular dementia, regardless of confounding factors (HR: 1.72, 95%CI 1.01-3.78; p = .048). Conversely, treatment with beta-blockers was not associated with increased risk of all-cause, Alzheimers and mixed dementia (HR:1.15; 95%CI 0.80-1.66; p = .44; HR:0.85; 95%CI 0.48-1.54; P = .59 and HR:1.35; 95%CI 0.56-3.27; p = .50, respectively). We observed that use of beta-blockers, as a class, is associated with increased longitudinal risk of vascular dementia in the general elderly population, regardless of cardiovascular risk factors, prevalent or incident history of atrial fibrillation, stroke, coronary events and heart failure. Further studies are needed to confirm our findings in the general population and to explore the mechanisms underlying the relationship between use of beta-blockers and increased risk of vascular dementia.