Physical activity, common brain pathologies, and cognition in community-dwelling older adults

Abstract

Objective: To examine the associations of physical activity, Alzheimer disease (AD), and other brain pathologies and cognition in older adults.

Methods: We studied 454 brain autopsies from decedents in a clinical-pathologic cohort study. Nineteen cognitive tests were summarized in a global cognitive score. Total daily physical activity summarized continuous multiday recordings of activity during everyday living in the community setting. A global motor ability score summarized 10 supervised motor performance tests. A series of regression analyses were used to examine associations of physical activity, AD, and other brain pathologies with global cognition proximate to death controlling for age, sex, education, and motor abilities.

Results: Higher levels of total daily activity (estimate 0.148, 95% confidence interval 0.053-0.244, SE 0.049, p = 0.003) and better motor abilities (estimate 0.283, 95% confidence interval, 0.175-0.390, SE 0.055, p < 0.001) were independently associated with better cognition. These independent associations remained significant when terms for AD and other pathologies were added as well as in sensitivity analyses excluding cases with poor cognition or dementia. Adding interaction terms, the associations of total daily activity and motor abilities with cognition did not vary in individuals with and without dementia. The associations of AD and other pathologies with cognition did not vary with the levels of total daily activity or motor abilities.

Conclusions: Physical activity in older adults may provide cognitive reserve to maintain function independent of the accumulation of diverse brain pathologies. Further studies are needed to identify the molecular mechanisms underlying this potential reserve and to ensure the causal effects of physical activity.

Figure 1. The association of total daily physical activity and motor abilities

Scatterplot showing the relationship of total daily physical activity and motor abilities in participants in these analyses with a locally weighted regression smooth function that is based on weighted locally linear fits.

Figure 2. The association of total daily physical activity and motor abilities with cognition before and after controlling for brain pathologies

(A) Illustration of the association of residual cognition with motor abilities controlling for age and sex (red line average regression line, and individual data points red circles) and after controlling for demographics and brain pathologies (blue line average regression line, and individual data points blue circles). Dashed line shows a line with a slope = 0 if there was no association between 2 predictors. (B) Model generated figure showing residual cognition and total daily physical activity after controlling for age and sex (red line average regression line, and individual values red circles) and after adding indices for 10 brain pathologies (blue line average regression line, and individual values blue circles). To highlight the association between motor function and cognition in both figures, a dashed line shows a line with a slope = 0 as would be seen if there was no association between 2 predictors. Blue lines in both figures show a significant association between higher total daily physical activity and higher motor abilities even after controlling for demographics and indices of 10 brain pathologies.