Physical Activity in Mid- and Late Life Linked to Markedly Lower Risk of Dementia

Higher levels of physical activity in midlife and late life were associated with a 36% to 45% lower risk of all-cause dementia in a long-term analysis of the Framingham Heart Study Offspring cohort published today in JAMA Network Open. Activity in midlife (ages 45-64) was associated with a 40% reduction in dementia risk over 26 years for individuals in the top 2 activity quintiles (HR 0.60, 95% CI 0.41-0.89; HR 0.59, 95% CI 0.40-0.88).

©miss irine/stock.adobe.com

Activity in late-life (ages 65-88) showed similar associations over 15 years (HR 0.64, 95% CI 0.42-1.00; HR 0.55, 95% CI 0.35-0.87). In contrast, physical activity in early adulthood did not correlate with incident dementia. Findings were consistent for Alzheimer disease and suggest that midlife and late-life may represent critical intervention windows for dementia prevention, according to lead author Phillip Hwang, PhD, MPH, assitant professor in the department of epidemiology at the Boston University School of Public Health and colleagues..

There are no curative treatments for dementia, and pharmacologic interventions for Alzheimer disease offer limited benefit and can carry adverse effects. As a result, researchers continue to examine modifiable risk factors across the life course. The 2024 Lancet Commission reported that up to 45% of dementia cases could be preventable by addressing 14 recognized risk factors, including physical inactivity in midlife. Prior research supports associations between physical activity and brain volume, connectivity, and preserved cognitive function, but uncertainty remains about when during adulthood activity confers the greatest protection.

Hwang and colleagues used prospective data from 4,354 adults in the Framingham Heart Study Offspring cohort. Participants were dementia-free at baseline and had self-reported physical activity measured in early adulthood (ages 26–44; n=1,526), midlife (45–64; n=1,943), or late life (65–88; n=885). Activity scores came from the physical activity index, which weights hours spent in sleep, sedentary behavior, and slight, moderate, or heavy activity.

Participants were followed for a mean of 37.2 years (early adulthood group), 25.9 years (midlife), and 14.5 years (late life). Over the study period, 567 individuals developed all-cause dementia: 62 in the early adult group (4%), 273 in midlife (14%), and 232 in late life (26%).

Models adjusted for age, sex, education, BMI, smoking status, hypertension, diabetes, hyperlipidemia, and APOE ε4 status.

FINDINGS

Physical activity in early adulthood showed no association with dementia risk. In midlife and late life, however, activity levels in the fourth and fifth quintiles consistently predicted lower risk of all-cause dementia and Alzheimer dementia.

For midlife activity, only the highest two quintiles were linked to lower dementia risk. Quintile 4 activity was linked with a 40% lower risk (HR 0.60, 95% CI 0.41-0.89) and quintile 5 with a 41% lower risk (HR 0.59, 95% CI 0.40-0.88) compared with quintile 1. Late-life activity showed a similar pattern: individuals in quintile 4 had a 36% lower risk (HR 0.64, 95% CI 0.42-1.00), and those in quintile 5 had a 45% lower risk (HR 0.55, 95% CI 0.35-0.87).

Activity intensity mattered in midlife. Moderate activity in quintile 4 or 5 was associated with 35% to 38% lower dementia risk (HR 0.65, 95% CI 0.44-0.97; HR 0.62, 95% CI 0.42-0.92). Heavy activity in quintile 5 also showed benefit (HR 0.66, 95% CI 0.44-1.00). In late life, however, the effect did not differ by activity intensity.

Associations varied by APOE ε4 status. Midlife physical activity reduced dementia risk only among ε4 noncarriers, whereas late-life activity showed benefit in both carriers and noncarriers.

Biological Mechanisms

Proposed mechanisms linking physical activity to cognitive health include increased cerebral blood flow, reduced systemic inflammation, enhanced neuroplasticity, and improved glucose metabolism. Exercise also increases the expression of brain-derived neurotrophic factor in animal models, supporting hippocampal neurogenesis and synaptic resilience.

Physical activity was self-reported at a single time point in each life period and may have been misclassified. The early adult group had fewer dementia events, which limits inference about early-life exposure. Participants were mostly of European ancestry, reducing generalizability. Higher mortality in lower-activity groups may have attenuated associations.

Hwang and colleagues conclude that midlife and late-life physical activity represent important windows for dementia risk reduction. The findings build life-course evidence that supports recommending moderate or higher levels of physical activity in these periods as part of broader prevention strategies.