Emotional and Behavioural Problems in Children with Neurodevelopmental Disorders: Using Cross-Disorder Phenotypes to Identify Informative Subgroups

Background: Autism Spectrum Disorder (ASD), Attention Deficit Hyperactivity Disorder (ADHD), and Obsessive Compulsive Disorder (OCD) are childhood-onset neurodevelopmental disorders (NDDs) linked to deficits in brain development and function. Accumulating evidence suggests there is notable clinical and genetic overlap among the three NDDs, lending support to the idea of common, cross-disorder aetiologies. Despite the evidence, our current psychiatric classification paradigm continues to treat these NDDs as distinct diagnostic entities. Such a categorical approach may be limiting our research to identify homogeneous subgroups of children who share similar profiles, irrespective of their primary NDD diagnosis. An alternative approach that identifies dimensional phenotypes that cut across conventional NDD diagnostic categories can inform treatment processes and the search for common cross-disorder aetiologies. Two clinical phenotypes in NDDs that warrant close attention and may be suitable for such cross-disorder investigations are emotional (internalizing) and behavioural (externalizing) problems.

Objectives: To use the phenotypes of emotional and behavioural problems to identify informative subgroups of children across conventional NDD diagnostic categories (ASD/ADHD/OCD).

Methods: The sample consisted of 784 children with NDDs (319 ASD, 329 ADHD, 136 OCD; 74.9% males; mean age: 10.7 years) participating in the Province of Ontario Neurodevelopmental Disorders (POND) Network project. To identify subgroups, data on the Child Behavior Checklist (CBCL 6-18) syndrome scales – Anxious/Depressed, Withdrawn/Depressed, Somatic Complaints, Social Problems, Thought Problems, Attention Problems, Rule-Breaking Behavior, and Aggressive Behavior – were used in Hierarchical Clustering analysis. The optimal number of clusters was selected using the scree-plot criterion. Derived clusters were characterized using sex ratio, the child’s age, social communication deficits (Social Communication Questionnaire; SCQ), obsessive compulsive behaviours (Toronto Obsessive Compulsive Rating Scale; TOCS) and adaptive behaviours (Adaptive Behavior Assessment System; ABAS-2)

Results: A 2–cluster solution provided the best fit to the data. Compared to those in Cluster 1, children in Cluster 2 (43% of sample) scored higher (within borderline or clinical range) on all CBCL syndrome scales (p< .001 for all) and total problems score (mean: 71.6 compared to 57.4; p<.001). The two clusters did not differ in terms of sex ratio or the child’s age. The proportion of children from each primary diagnostic category in Cluster 2 was 45.5% of ASD, 45.9% of ADHD, and 30.1% of OCD. Within Cluster 2, children with ASD scored higher on the SCQ (mean ASD: 22.2 > ADHD: 10.2 and OCD: 7.7; p<.001) and had the lowest scores on the ABAS (mean ASD: 60.7 < ADHD: 75.7 and OCD: 80.1; p<.001). Children with OCD had the highest scores on the TOCS (mean OCD: 20.6 > ASD: -11.6 > ADHD: -19.9; p<.001).

Conclusions: Study findings suggest that emotional and behavioural problems are useful phenotypes for the identification of informative subgroups of children with NDDs that may benefit from cross-disorder treatments. At the same time, notable within-subgroup variability suggests that primary diagnostic categories (ASD/ADHD/OCD) provide useful information on unique phenotypes (e.g., social communication deficits, obsessive compulsive behaviors, and adaptive behaviors) requiring disorder-specific treatments. This work has the potential to inform research on the shared and unique aetiologies in NDDs.