Identifying ADHD Subtypes Based on Subclinical Autistic Traits, Behavioral and Emotional Symptoms, and Executive Function

Transdiagnostic dimensional approaches support characterizing mental disorders through dimensional perspectives. Based on this recommendation, we subtyped children with ADHD using autistic and ADHD traits, cognitive and psychopathological domains. Our goal was to identify distinct and meaningful subgroups within the ADHD population. We used 892 ADHD participants aged 5 to 18 from the HBN data set. Using the Latent Profile Analysis (LPA), we identified subgroups based on the following features: (1) Inattentive, Hyperactive–Impulsive subscales measured by SWAN (Strengths and Weaknesses of ADHD-symptoms and Normal-behaviours), (2) cognitive flexibility, working memory, processing speed, and inhibitory control assessed with NIH Toolbox Cognition Battery, (3) autistic traits evaluated with SRS-2 (Social Responsiveness Scale-Second Edition), SCQ (Social Communication Questionnaire), and ASSQ (Autism Spectrum Screening Questionnaire), (4) internalizing symptoms assessed with the MFQ-P (Mood and Feelings Questionnaire) and the SCARED-P (Screen for Child Anxiety Related Emotional Disorders-Parent Report), (5) externalizing symptoms measured with the Rule-Breaking Behavior and Aggressive Behavior subscales of the CBCL (Child Behavior Checklist). All classes’ features showed significant results in differentiating class profiles, with externalizing symptoms showing the largest effect size and inattention, working memory, anxiety, and processing speed showed the smallest (but still significant). We found four meaningful classes: a moderately high EF/ very low symptoms profile, a high EF/ mild symptoms profile, a moderately low EF/ high symptoms profile, and an impaired EF/ low symptoms profile. We used the PSI-4 (Parenting Stress Index, Fourth Edition), SDSC (Sleep Disturbance Scale for Children), PCIAT (Parent-Child Internet Addiction Test), CGAS (Children’s Global Assessment Scale), and WIAT (Wechsler Individual Achievement Test) for external validation of the classes. The classes with the worst EF performance had lower academic achievement, and classes with more severe symptoms had greater internet addiction, parenting stress, and sleep disturbance problems. The combination of the high EF and low symptoms yielded the best outcome.