Objective: The objective of this study was to evaluate cross-sectional and longitudinal associations between ankle-brachial index (ABI) and indicators of cognitive function. Design: Randomized clinical trial (Lifestyle Interventions and Independence for Elders Trial). Setting: Eight US academic centers. Participants: A total of 1601 adults ages 70-89 years, sedentary, without dementia, and with functional limitations. Measurements: Baseline ABI and interviewer- and computer-administered cognitive function assessments were obtained. These assessments were used to compare a physical activity intervention with a health education control. Cognitive function was reassessed 24 months later (interviewer-administered) and 18 or 30 months later (computer-administered) and central adjudication was used to classify individuals as having mild cognitive impairment, probable dementia, or neither. Results: Lower ABI had a modest independent association with poorer cognitive functioning at baseline (partial r= 0.09; P < .001). Although lower baseline ABI was not associated with overall changes in cognitive function test scores, it was associated with higher odds for 2-year progression to a composite of either mild cognitive impairment or probable dementia (odds ratio 2.60 per unit lower ABI; 95% confidence interval 1.06-6.37). Across 2 years, changes in ABI were not associated with changes in cognitive function. Conclusion: In an older cohort sedentary individuals with dementia and with functional limitations, lower baseline ABI was independently correlated with cognitive function and associated with greater 2-year risk for progression to mild cognitive impairment or probable dementia.