BACKGROUND AND PURPOSE: We aimed to study if physical activity could interfere with progression for cognitive impairment and dementia in older people with white matter changes living independently. METHODS: The LADIS (Leukoaraiosis and Disability) prospective multinational European study evaluates the impact of white matter changes on the transition of independent elderly subjects into disability. Subjects were evaluated yearly during 3 years with a comprehensive clinical protocol and cognitive assessment with classification of cognitive impairment and dementia according to usual clinical criteria. Physical activity was recorded during the clinical interview. MRI was performed at entry and at the end of the study. RESULTS: Six hundred thirty-nine subjects were included (74.1±5 years old, 55% women, 9.6±3.8 years of schooling, 64% physically active). At the end of follow-up, 90 patients had dementia (vascular dementia, 54; Alzheimer disease with vascular component, 34; frontotemporal dementia, 2), and 147 had cognitive impairment not dementia. Using Cox regression analysis, physical activity reduced the risk of cognitive impairment (dementia and not dementia: β=-0.45, P=0.002; hazard ratio, 0.64; 95% CI, 0.48-0.85), dementia (β=-0.49, P=0.043; hazard ratio, 0.61; 95% CI, 0.38-0.98), and vascular dementia (β=-0.86, P=0.008; hazard ratio, 0.42; 95% CI, 0.22-0.80), independent of age, education, white matter change severity, medial temporal atrophy, previous and incident stroke, and diabetes. CONCLUSIONS: Physical activity reduces the risk of cognitive impairment, mainly vascular dementia, in older people living independently.