Clinical, physical and lifestyle variables and relationship with cognition and mood in aging: a cross-sectional analysis of distinct educational groups

It is relevant to unravel the factors that may mediate the cognitive decline observed during aging. Previous reports indicate that education has a positive influence on cognitive performance, while age, female gender and, especially, depressed mood were associated with poorer performances across multiple cognitive dimensions (memory and general executive function). Herein, the present study aimed to characterize the cognitive performance of community-dwelling individuals within distinct educational groups categorized by the number of completed formal school years: "less than 4," "4, completed primary education," and "more than 4." Participants (n = 1051) were randomly selected from local health registries and representative of the Portuguese population for age and gender. Neurocognitive and clinical assessments were conducted in local health care centers. Structural equation modeling was used to derive a cognitive score, and hierarchical linear regressions were conducted for each educational group. Education, age and depressed mood were significant variables in directly explaining the obtained cognitive score, while gender was found to be an indirect variable. In all educational groups, mood was the most significant factor with effect on cognitive performance. Specifically, a depressed mood led to lower cognitive performance. The clinical disease indices cardiac and stroke associated with a more negative mood, while moderate increases in BMI, alcohol consumption and physical activity associated positively with improved mood and thus benefitted cognitive performance. Results warrant further research on the cause-effect (longitudinal) relationship between clinical indices of disease and risk factors and mood and cognition throughout aging.This work was funded by the European Commission (FP7): "SwitchBox" (Contract HEALTH-F2-2010-259772). Nadine C. Santos is supported by a SwitchBox post-doctoral fellowship; Carlos Portugal-Nunes and Liliana Amorim by a MyHealth (DoIT-Desenvolvimento e Operacionalizacao da Investigacao de Translacao, Contract 13853) research and PhD fellowships, respectively. We are thankful to all study participants. The authors would like to acknowledge all colleagues who assisted with participant recruitment and evaluation. We thank Dr. Matthew Stump and Daniel Zimmerman for the careful reading of the manuscript