Long-term physical activity, brain structure, and cognitive function: a systematic review of longitudinal observational evidence

Background Physical inactivity is a key modifiable risk factor for dementia, and physical activity is proposed to confer neuroprotective effects on brain structure and cognitive function. Objectives To synthesize longitudinal observational evidence on associations between long-term habitual physical activity and brain structural outcomes and cognitive function across adult age groups. Methods Five databases were searched from inception to August 2024. Eligible studies enrolled adults aged ≥ 18 years without cognitive impairment or dementia, used validated accelerometry or self-report physical activity assessment, and employed longitudinal observational designs with a minimum 12-month exposure-to-outcome window. Risk of bias was assessed using the ROBINS-E tool. Results Twenty studies were included in the review. Higher habitual physical activity was consistently associated with preserved hippocampal and prefrontal grey matter volume and attenuated white matter microstructural decline. Executive function and processing speed showed the most consistent cognitive associations, with evidence of a dose-response accumulation effect across adulthood. Associations appeared amplified in APOE-ε4 carriers. Bidirectional analyses indicated preserved brain structure also predicted subsequent physical activity, suggesting reverse causation cannot be excluded. All 20 studies were rated “some concerns” on the ROBINS-E scale. Conclusion Longitudinal observational evidence is consistent with the hypothesis that sustained habitual physical activity is associated with preserved brain structure and attenuated cognitive decline, though causal inference is not warranted given the limitations of the observational design.