No association between gain in body mass index across the life course and midlife cognitive function and cognitive reserve-The 1946 British birth cohort study

Background: The association between lifelong body mass index (BMI) and cognitive function has not been comprehensively studied. Methods: In more than 2000 men and women born in 1946, we tested associations between BMI gain at 15, 20, 26, 36, 43, and 53 years with respect to the previous measure (gain at age 15 years with respect to BMI at age 11 years), and semantic fluency (animal naming) and cognitive reserve (the National Adult Reading Test) at age 53 years, and verbal memory (word list recall) and speed/concentration (letter cancellation) at ages 43 and 53 years. Measures of BMI gain were adjusted in stages for childhood intelligence, education, socioeconomic position (SEP), lifestyle, and vascular risk factors. Results: Independent of childhood intelligence, BMI gain between ages 26 and 36 years was associated with lower memory scores (beta per SD increase in BMI in men = -0.11; 95% confidence interval [CI]: -0.19, -0.02), verbal fluency (beta in women = -0.11; 95% CI: -0.20, -0.02), and lower National Adult Reading Test score (beta in women = -0.08; 95% CI: -0.15, -0.01), but not with speed/concentration (beta in men = 0.02; 95% CI: -0.11, 0.07). Associations were largely explained by educational attainment and SEP (P >= .10). However, BMI gain at 53 years in men was independently associated with better memory (beta = 0.12; 95% CI: 0.03, 0.22), and both underweight (beta = -1.54; 95% CI: -2.52, -0.57) and obese (beta = -0.30; 95% CI: -2.52, -0.57) women at 53 years had significantly lower memory scores. Conclusion: The adverse effect of higher BMI gain on midlife cognitive function and cognitive reserve is independent of childhood intelligence but not of education and SEP. The independent association between greater BMI gain in midlife and better cognitive function deserves further investigation. (C) 2012 The Alzheimer's Association. All rights reserved.