Attention problems and cortical maturation in a large longitudinal sample of youths: The importance of accounting for sex differences

Abstract

Delayed age-related cortical thinning has been proposed as a biomarker of attention-related psychopathology and attention-deficit/hyperactivity disorder (ADHD), but these findings have not been adequately replicated in large, longitudinal samples with sufficient power to account for potential confounds such as co-occurring psychopathology and sex differences in neurodevelopment. Here, we examined whether previously reported associations between attention problems and delayed cortical thickness development could be replicated in a large, longitudinal cohort of youths. Leveraging data from the Adolescent Brain Cognitive Development Study, linear mixed-effects models were used to assess the association between parent-reported attention problems (AP) on the Child Behavior Checklist and age-related cortical thickness change (N = 26,496 MRI scans from 11,025 unique participants). Secondary analyses examined the association between a polygenic risk score for ADHD and cortical thickness development. In initial sex-pooled analyses, we observed associations between AP and reduced rates of cortical thinning (β = 00594 to 0.0145, FDR-adjusted P < 0.05). However, when an age × sex interaction term was included in the model, these associations were no longer evident (all FDR-adjusted P > 0.05 across ROIs). Follow-up sex-stratified analyses revealed no significant age × AP interactions on cortical thickness in males or females. Further, there was no evidence of genetic liability for ADHD being associated with reduced age-related cortical thinning. Taken together, these findings suggest that previously reported associations between AP and delayed cortical thinning likely reflect unaccounted-for sex differences in neurodevelopment rather than AP-specific maturational delays, questioning the utility of cortical maturation patterns as diagnostic biomarkers for ADHD-related behavior.