5,000 Steps May Protect Your Brain Against Alzheimer's Progression
Harvard researchers tracked 296 older adults for 14 years and found that moderate walking slows cognitive decline.
A landmark 14-year study from Harvard Medical School has identified a precise threshold of daily physical activity that may delay Alzheimer‘s-related cognitive decline by up to seven years in at-risk older adults. Published in Nature Medicine in November 2025, the research provides the first longitudinal evidence that moderate walking—between 5,000 and 7,500 steps per day—significantly slows the accumulation of tau proteins in the brain, the toxic tangles directly responsible for neurodegeneration in Alzheimer’s disease.
The study tracked 296 cognitively unimpaired adults aged 50 to 90 using objective pedometer measurements and serial brain imaging with PET scans, revealing that physical activity does not reduce amyloid plaques but instead interrupts the downstream cascade from amyloid to tau to cognitive symptoms. Critically, the protective effect was most pronounced in sedentary individuals with elevated brain amyloid—the group at highest risk for progression—with even modest increases from 3,000 to 5,000 daily steps associated with a 40% reduction in cognitive decline over nine years compared to inactive participants. The findings suggest that we don’t need to hit the popular “10,00 step target”, and instead, this more achievable goal may better help deliver cognitive protection for older adults.
Aim
The aim was to determine whether objectively measured physical activity is associated with slower progression of Alzheimer’s disease pathology and whether these biological changes mediate the relationship between physical activity and cognitive decline in preclinical Alzheimer’s disease.
Methods
Researchers recruited 296 cognitively unimpaired adults aged 50-90 years (mean age 72.3 years, 59% female, mean education 15.8 years) who were part of the Harvard Aging Brain Study. Participants wore waistband-mounted pedometers for seven consecutive days to objectively measure daily step counts, with a mean of 5,719 steps per day recorded.
Brain imaging was conducted using Pittsburgh compound-B (PiB) PET scans to measure global amyloid-beta burden (n=241 with longitudinal data) and 18F-Flortaucipir PET scans to measure tau protein in the inferior temporal cortex (n=172 with longitudinal data). Cognition was assessed annually using the Preclinical Alzheimer’s Cognitive Composite-5 (PACC5), and functional decline was measured using the Clinical Dementia Rating Sum of Boxes (CDR-SOB).
Linear mixed effects models examined the interactive effects of baseline physical activity and amyloid burden on longitudinal tau accumulation, cognitive decline, and functional decline, adjusting for age, sex, APOE ε4 status, and education. Moderated mediation analyses determined whether tau accumulation mediated the relationship between physical activity and cognitive outcomes.
Results
Physical activity showed no association with longitudinal amyloid-beta accumulation. However, there was a significant interaction between baseline physical activity and amyloid burden on longitudinal tau accumulation, whereby higher physical activity was associated with slower amyloid-related inferior temporal cortex tau accumulation.
Higher physical activity was associated with slower cognitive decline as measured by PACC5 in individuals with elevated amyloid, and slower functional decline as measured by CDR-SOB.
Mediation analysis revealed that slower tau accumulation fully mediated the association between higher physical activity and slower cognitive decline in individuals with elevated amyloid, and partially mediated the association with functional decline. Dose-response analyses using ordinal activity levels (inactive: ≤3,000 steps; low: 3,001-5,000 steps; moderate: 5,001-7,500 steps; active: ≥7,501 steps) showed that even low levels of activity were associated with 34-40% slower cognitive and functional decline over 9 years compared to inactive individuals. The protective associations plateaued at moderate activity levels (5,001-7,500 steps per day), with no additional benefit observed at higher activity levels.
For individuals with elevated amyloid, estimated PACC5 z-scores declined by 2.5 points (inactive), 1.5 points (low activity), 1.1 points (moderate activity), and 1.2 points (active) from baseline to 9 years. Inactive individuals with elevated amyloid reached the threshold for cognitive impairment (-1.5 PACC5 z-score) at 6.5 years, while those with low, moderate, and active levels reached it at 9.6, 13.6, and 12.7 years, respectively.
Book a call with me here.
We’ll sit down and upgrade or build out your entire training and performance program in one session.