Subgrouping school-aged children on the autism spectrum based on co-occurring psychiatric symptoms

Background: Phenotypic heterogeneity along the autism spectrum is compounded by co-occurring psychiatric conditions. Deriving subgroups of autistic individuals based on symptoms of these conditions could better our understanding of psychiatric symptom patterns existing within this population. This study's goals were to derive subgroups of school-aged autistic children based on co-occurring psychiatric symptoms while controlling for age and sex and to examine correlates of subgroup membership while controlling for the degree of ASD-related diagnostic features. Method: Latent class models were estimated in a sample from the Simons Simplex Collection (n = 2087) using "borderline/clinical" versus "normative" range data from five of the DSM-Oriented Scales from the CBCL/6-18 as indicator variables. We evaluated the predictive value of NVIQ < 70, atypical sleep duration, allergies/autoimmune conditions, gastrointestinal conditions, and neurological conditions on subgroup membership using multinomial logistic regression. Results: Four subgroups emerged: Low Psychiatric Symptoms (41.02%), Externalizing Symptoms (12.36%), Internalizing Symptoms (31.58%), and High Psychiatric Symptoms (15.05%). Key findings were that NVIQ < 70 was associated with decreased odds of belonging to the Internalizing Symptoms and High Psychiatric Symptoms subgroups over the Low Psychiatric Symptoms subgroup, while atypical sleep duration and gastrointestinal conditions were associated with increased odds of belonging to the Externalizing Symptoms and High Psychiatric Symptoms subgroups. Neurological conditions were also associated with increased odds of belonging to the Externalizing Symptoms subgroup. Conclusion: Distinct patterns of psychiatric symptoms exist within school-aged autistic children and are correlated with NVIQ < 70, atypical sleep duration, and medical conditions, providing insights for clinical practice and etiology-driven research.