The Association Between Physical Activity and Domain-Specific Cognitive Function in the Elderly: A Cross-Sectional...

Population aging is intensifying worldwide, increasing the prevalence of dementia. More than 50 million people globally are affected by dementia. There is some evidence that physical activity (PA) benefits cognitive function (CF). However, it is unclear which types and amounts of PA are best for specific cognitive domains.

This study used a cross-sectional design and genetic analysis to test its hypotheses. Participants were drawn from the National Health and Nutrition Examination Survey (NHANES) 2011-2014, including 2578 adults aged 60 years or older. Cognitive abilities were measured using standardized tests, and PA levels were assessed with the Global Physical Activity Questionnaire (GPAQ). The analysis included multivariable regression, threshold effect testing, and subgroup analysis to explore links between PA patterns, intensities, and specific CF domains. Furthermore, a Mendelian randomization (MR) approach was employed to assess whether PA intensity exerts an effect on CF.

NHANES data analysis showed that the regular activity (RA) pattern was positively associated with the Consortium to Establish a Registry for Alzheimer's Disease Word Learning test (CERAD-WL) scores (β = 0.07; 95% confidence interval [CI]: 0.02, 0.16; p = 0.018), Animal Fluency Test (AFT) scores (β = 0.31; 95% CI: 0.21, 0.41; p = 0.002), Digit Symbol Substitution Test (DSST) scores (β = 0.19; 95% CI: 0.09, 0.30; p = 0.017), and overall CF (β = 0.23; 95% CI: 0.14, 0.32; p < 0.004). Threshold-effect analysis revealed an inverted U-shaped relationship between PA levels and cognitive performance. The inflection points occurred at 650, 535, and 550 min per week for AFT, DSST, and overall CF, respectively. Above these values, further cognitive gains plateaued. The inverse-variance weighted (IVW) method of the MR analyses showed that moderate-intensity physical activity (MPA) increased correct counting (odds ratio [OR] = 0.710; 95% CI: 0.606, 0.832; p < 0.001), fluid intelligence (OR = 0.470; 95% CI: 0.346, 0.638; p < 0.001), and overall cognitive performance (OR = 0.707; 95% CI: 0.653, 0.764; p < 0.001). For vigorous-intensity physical activity (VPA), we observed causal associations with memory outcomes, reaction time, correct matches, fluid intelligence, and cognitive performance.

The findings suggest that PA is associated with multiple domains of CF, and the RA pattern is linked to better cognitive performance. Genetically, VPA appears to have a stronger promotive effect. Thus, different patterns and intensities of PA may help delay cognitive decline and maintain brain health in older adults.