The Relation between Traumatic Brain Injury and Neuropsychological Functioning: Sociodemographic Moderators and Biopsychosocial and Behavioral Mediators

Abstract

Traumatic brain injury (TBI) has been shown to have known acute cognitive effects, but less is known about more distal cognitive correlates and whether these vary as a function of sociodemographic factors. Additionally, even less is known about the potential biopsychosocial and behavioral mediators of these relations. Thus, the purpose of this study was, first, to examine whether sex and race moderate the relations between TBI history and cognition. We further evaluated clusters of biopsychosocial and behavioral variables as potential mediators of theses associations. Participants were 250 adults with a self-reported history of TBI (62% male, 57% White, 50% above the 125% poverty line, mean age = 47.6 years, mean education 12.2 years) and 500 age-matched adults without a history of TBI (61% male, 63% White, 36% above the 125% poverty line, mean age = 47.0 years, mean education 12.5 years) from the Healthy Aging in Neighborhoods of Diversity Across the Life Span (HANDLS) Study. TBI history and sociodemographic information (age, sex, race, poverty status, and education) were self-reported by participants. Cognitive function was measured across a variety of domains including verbal memory (California Verbal Learning Test Total Recall, Learning Curve, Short Delay Free Recall, and Long Delay Free Recall), visual memory (Benton Visual Retention Test), perceptuo-motor speed and manual dexterity (Trail Making Test, Part A), attention/working memory (Digit Span Forward and Backward), language/executive function (Category Fluency), and executive function (Trail Making Test, Part B). Biopsychosocial and behavioral clusters were identified using principal components analysis for the biological (systolic blood pressure, fasting glucose level, and body mass index), psychological (post-traumatic stress disorder symptomatology, depressive symptomatology, and trait anger), social (emotional and instrumental support), and behavioral (alcohol status, smoking status, and a composite drug use variable) domains. Mixed-effect models were used to examine the interaction of sex and TBI history on a variety of neuropsychological measures covarying for education, poverty status, and race. These models were repeated for the interaction of race and TBI history, and a 3-way interaction of sex, race, and TBI history. For each significant interaction, the biopsychosocial and behavioral components were incrementally added to each model as potential explanatory factors. Results showed that African-Americans with a history of TBI performed significantly worse on CVLT total, a measure of verbal memory (F (1, 695) = 6.161, p = .013), than African-Americans without a TBI history, and that men with a TBI history performed significantly worse on Digits Forward and Backward, measures of attention (F (1, 480.902) = 4.045, p = .045) and working memory (F (1, 478.600) = 5.952, p = .014), respectively. The relation of race and TBI history to memory was partially mediated by the cluster of psychological variables, and the relation of sex and TBI history to attention was partially mediated by the biological cluster. The sex, TBI history, and working memory association was not significantly mediated by any cluster. Overall, these findings suggest that distal cognitive effects, particularly verbal memory and attention/working memory, from TBI may be most pronounced in African-Americans and men, and these relations may be mediated by psychological and biological factors respectively. However, further exploration is needed to clarify these relations. These biopsychosocial factors should be an important focus in clinical practice, particularly for African-Americans and men, to aid in the recovery of cognitive deficits.