BACKGROUND: The relationship between stroke risk and cognitive ageing has not been adequately studied in non-elderly populations. The objective of this study was to examine whether 10-year risk for incident stroke is associated with cognitive decline. METHODS: Study sample comprised of 4512 men and 1741 women, mean age 55.6 years, from the Whitehall II study, a longitudinal British cohort study. The Framingham Stroke Risk Profile was used to assess 10-year risk of stroke. It incorporates age, systolic blood pressure, diabetes mellitus, smoking status, prior cardiovascular disease, atrial fibrillation, left ventricular hypertrophy, and use of hypertensive medication. Measures of cognitive function consisted of tests of reasoning, memory, phonemic and semantic fluency, and vocabulary, assessed three times over 10 years. Linear mixed models were used to determine longitudinal associations between stroke risk and subsequent cognitive decline over 10 years. RESULTS: Higher stroke risk at baseline was associated with faster rate of cognitive in tests of reasoning, verbal fluency, vocabulary and global cognition. For example, compared to persons in the low stroke risk group (<2.5%), those in the moderate stroke risk group (2.5≤ stroke risk <5%) and the high stroke risk group (≥5%) had a 12.5% and 43.8% faster rate of decline in, phonemic fluency respectively. CONCLUSIONS: Higher 10-year stroke risk in middle age is associated with faster rate of cognitive decline in more than one cognitive domain. These results support early targeting of vascular risk factors to prevent or delay cognitive decline.