Body fat and the cognitive pattern: A population-based study

OBJECTIVE: The relationship between body fatness and cognitive pattern at a population level was investigated. METHODS: Among 500 unselected subjects from the general population, the role of body mass index (BMI) and body fat mass (BFM) on a mini-mental state examination (MMSE) and on a battery of paper and pencil neuropsychological tests was analyzed. Multiple linear regressions, accounting for potential confounders, were used. RESULTS: In fully adjusted models, MMSE (coefficient +0.027, 95% confidence intervals, 0.017-0.177), the clock drawing test (+0.141, 0.053-0.226), and the trail making test A (+1.542, 0.478-2.607) were positively associated with BMI. Adding BFM to the models, no associations were observed. The tests were also positively associated with BFM (+0.056, 0.021-0.091; +0.063, 0.025-0.101; +0.592, 0.107-1.077; respectively). At analysis of covariance, the same tests were significantly better performed over 29.4 kg m(-2) of BMI. After adding BFM as further confounder, all differences in performance across BMI were no longer significant. The three tests were better performed over 34.6 kg of BFM. CONCLUSIONS: Higher BMI and particularly higher BFM are positively associated with better performance at the cognitive tasks exploring selective attention and executive functions.