Hypertension has been associated with subclinical injury in the brain and may therefore increase the impact of an incident stroke on cognitive function. The Study of Osteoporotic Fractures (SOF) is a prospective, observational study of 9,704 women aged 65 years and older recruited from four U.S. metropolitan areas. Blood pressure was measured at study entry, and cognitive decline was defined by the change from prestroke to poststroke cognitive testing. During an average follow-up of 6.8 years, incident stroke occurred in 260 participants (3.1%) who had previously completed baseline cognitive testing. Among participants with stroke, 119 completed follow-up cognitive testing a median of 1.9 years after the stroke, 80 died before the next study visit, and 61 did not complete further cognitive testing. After adjustment for demographic factors and other confounders, pre-existing hypertension was a strong predictor of cognitive decline when a stroke occurred (odds ratio [OR], 4.07; 95% confidence interval [CI], 1.37-12.1). In contrast, hypertension was only weakly associated with cognitive decline in the absence of stroke (OR, 1.13; 95% CI, 1.04-1.22) (p for interaction = 0.032). Pre-existing hypertension in women is associated with a greater impact of stroke on cognitive function, possibly by influencing the ability to tolerate or recover from brain injury.