OBJECTIVES To explore the association between use of antipsychotics and risk of cerebrovascular accident (CVA) in individuals with dementia aged 65 and older. DESIGN Population-based case-control study. SETTING UK-based electronic primary care records in the General Practice Research Database (GPRD). PARTICIPANTS Individuals with dementia aged 65 and older registered in the database between January 1, 1995, and June 22, 2007. MEASUREMENTS Odds ratio (OR) of CVA in users versus nonusers of antipsychotics (typical or atypical) and in users of typical versus atypical antipsychotics. Multivariate analyses were performed using logistic regression models to adjust for potential confounders: demographic variables, comorbidity, and concomitant treatments. RESULTS After adjusting for confounding variables, the OR of CVA associated with use of only typical antipsychotics versus no antipsychotics in individuals with dementia aged 65 and older was 1.16 (95% confidence interval (CI)=1.07-1.27) and for use of only atypical antipsychotics versus no antipsychotics was 0.62 (95% CI=0.53-0.72). In the comparison of typical versus atypical antipsychotics, the OR was 1.83 (95% CI=1.57-2.14). CONCLUSION No reasons were found to question the cerebrovascular safety of atypical antipsychotics in older adults with dementia. The typical antipsychotics appear to be associated with a higher risk of CVA, although the risk disappears after use is discontinued.