Bidirectional Associations Between Adiposity and Cognitive Function and Mediation by Brain Morphology in the ABCD Study

IMPORTANCE: Most epidemiologic studies examine the brain as an outcome in relation to adiposity (ie, the brain-as-outcome perspective), but it is also a potential risk factor associated with adiposity accumulation over time (ie, the brain-as–risk factor perspective). The bidirectionality hypothesis has not been fully explored in adolescent samples previously. OBJECTIVE: To assess bidirectional associations between adiposity and cognitive function in youth and test mediational pathways through brain morphology (specifically the lateral prefrontal cortex [LPFC]), lifestyle behaviors, and blood pressure. DESIGN, SETTING, AND PARTICIPANTS: This cohort study uses data (wave 1-3; 2 years of follow-up) from the Adolescent Brain Cognitive Development (ABCD) Study, a long-term longitudinal investigation of brain development in the United States launched in 2015, which recruited 11 878 children aged 9 to 10 years at inception. Data analysis was performed from August 2021 to June 2022. MAIN OUTCOMES AND MEASURES: Multivariate multivariable regression analyses were used to assess bidirectional associations of indicators of cognitive function (eg, executive function, processing speed, episodic memory, receptive vocabulary and reading skills) and adiposity (eg, body mass index z scores [zBMI] and waist circumference [WC]). Mediators considered for this investigation were lifestyle variables (eg, diet and physical activity), blood pressure, and the morphology of the LPFC and its subregions. RESULTS: A total of 11 103 individuals (mean [SD] age, 9.91 [0.6] years; 5307 females [48%]; 8293 White individuals [75%] and 2264 Hispanic individuals [21%]) were included in the current study. Multivariate multivariable regression analyses revealed that higher baseline zBMI and WC were associated with worse follow-up episodic memory (β, −0.04; 95% CI, −0.07 to −0.01) and better vocabulary (β, 0.03; 95% CI, 0.002 to 0.06) task performance, in covariate adjusted models. Similarly, superior baseline executive function (zBMI: β, −0.03; 95% CI, −0.06 to −0.01; WC: β, −0.04; 95% CI, −0.07 to −0.01) and episodic memory (zBMI: β, −0.04; 95% CI, −0.07 to −0.02; WC: β, −0.03; 95% CI, −0.06 to −0.002) task performance were associated with better follow-up adiposity status in covariate adjusted models. Cross-lagged panel models with latent variable modeling had a bidirectional association with executive function task performance (brain-as-outcome: β, −0.02; 95% CI, −0.05 to −0.001; brain-as–risk factor: β, −0.01; 95% CI, −0.02 to −0.003). The hypothesized associations were statistically mediated by LPFC volume and thickness, physical activity, and blood pressure. CONCLUSIONS AND RELEVANCE: In this cohort study, executive function and episodic memory were bidirectionally associated with adiposity indices over time in this adolescent sample. These findings suggest that the brain can be both a risk factor and an outcome of adiposity; this complex bidirectional association should be taken into account in future research and clinical practice.