Physical activity and cognitive performance: cross-sectional analysis of the Brazilian longitudinal study of adult health (ELSA-Brazil).

INTRODUCTION: Evidence on the relationship between Leisure-Time Physical Activity (LTPA), Commuting Physical Activity (CPA), and cognitive performance is inconsistent. OBJECTIVE: To examine the associations between LTPA, CPA, and cognitive performance in adults from the Brazilian Longitudinal Study of Adult Health (ELSA‒Brazil). METHODS: This cross-sectional analysis included 13,340 participants (median age 51 years; 54.5% women). Physical Activity (PA) was assessed using the long-form International Physical Activity Questionnaire, including LTPA and CPA. Cognitive performance was measured by word memory, verbal fluency, and the Trail Making Test Part B. Z-scores were computed for each test, and a global cognitive score was derived as their mean. Linear regression models estimated associations between PA levels and cognitive performance after multivariable adjustment, including occupational status. Analyses were stratified by sex and conducted in the overall sample. RESULTS: LTPA inactivity was associated with lower scores for verbal fluency (β = -0.07; 95% CI: -0.10 to -0.04), executive function (β = -0.10; 95% CI: -0.13 to -0.06), and global cognition (β = -0.06; 95% CI: -0.08 to -0.03). CPA inactivity showed associations in the opposite direction, with higher executive function (β = 0.08; 95% CI: 0.05 to 0.12) and memory in men (β=0.05; 95% CI: 0.001 to 0.10). After adjustment for occupational status, CPA associations attenuated and lost significance. CONCLUSION: In this cross-sectional study, the relationship between PA and cognition differed by domain. LTPA was associated with better cognitive scores, while CPA showed non-favorable or null associations after adjustment.