Association of cognitive performance with overall, dosage, intensity, and domain physical activity in aging: NHANES 2011-2014

Background: The specific relationship between PA in detailed types and cognition is still unclear due to limited evidence. Our study aimed to investigate the relationship between cognitive performance and various aspects of physical activity, including overall activity, dosage, intensity levels [moderate physical activity (MPA), vigorous PA], and different domains of activity [occupational PA (OPA), transportation PA (TPA), and leisure-time PA (LTPA)] in older adults using data from the NHANES database.

Methods: This cross-sectional analysis used data from 2 cycles of NHANES (2011-2014). PA was determined through participants' self-reports using the Global Physical Activity Questionnaire (GPAQ). Cognitive performance was evaluated by the presence of psychometric mild cognitive impairment (p-MCI), identified based on a composite measure derived from three cognitive tests including the Consortium to Establish a Registry for Alzheimer's Disease (CERAD), the Animal Fluency test, and the Digit Symbol Substitution test (DSST). Logistic regression models were used to evaluate the association.

Results: A total of 2588 participants aged 60 years or older were included, with an average age of 69.4 years and 48% being male. In the fully adjusted model, compared to no PA, performing 300 min of PA, and MPA were associated with 44%, and 33% reductions in the prevalence of p-MCI, respectively. Additionally, engaging in 1-149 min/week [OR 0.56, 95% CI (0.33-0.92)] and ≥ 300 min/week [OR 0.66, 95% CI (0.44-0.96)] of OPA, as well as ≥ 300 min/week [OR 0.56, 95% CI (0.36-0.86)] of LTPA, were also associated with a lower prevalence of p-MCI. Additionally, engaging in a diverse range of PA had better outcomes.

Conclusions: Our results suggested a positive association between higher levels of PA and enhanced cognitive performance. Different intensities and domains of PA have varying impacts on cognition. Future exploration, such as objectively measured PA and longitudinal studies were needed to validate our conclusion.