Background Besides elevated mean systolic blood pressure (BP), BP variability, pulse pressure (PP), or mean arterial pressure (MAP) could be important targets for dementia prevention. This study investigated these associations in participants with stroke or transient ischemic attack from the Perindopril Protection Against Recurrent Stroke Study (PROGRESS) trial.Methods BP parameters were calculated from measurements for 18 months on six occasions. Risk of cognitive decline according to Mini-Mental State Examination scores and/or adjudicated dementia outcomes per 1-standard deviation higher for each BP parameter was calculated using modified Poisson regression models with a log link. Subgroup analysis was conducted by mild cognitive impairment (MCI) at baseline, sex, and stroke subtype.Results Of 5128 participants recruited between May 1995 and November 1997, 567 (11.1%) developed cognitive decline/dementia after 4 (median) years of follow-up until 2001. For systolic BP (SBP), higher mean, maximum, cumulative load, and last measure were associated with higher risk of cognitive decline/dementia, but mean conferred the highest risk (relative risk 1.18, 95% confidence intervals [CIs] 1.06-1.31). For diastolic BP, higher mean, maximum, variability, cumulative load, slope, and last measure were associated, while load showed the strongest association (1.23, 95% CI 1.12-1.36). Only the mean PP was associated (1.11, 95% CI 1.00-1.24). For MAP, MAP load showed the strongest association (1.20, 95% CI 1.09-1.33). Systolic, diastolic, and MAP variability showed a higher risk in those without MCI.Conclusions Besides mean SBP, higher variability, cumulative load, slope, and last measures of diastolic BP, PP, and MAP were associated with increased risk of cognitive decline/dementia.Clinical Trial Registration This trial was not registered because enrollment began before July 1, 2005.