Longitudinal associations between autistic children's anxiety and social communication differences: The moderating role of executive function behaviours.

This is an observational study, so it cannot show that anxiety or social communication differences caused the other. The sample was recruited with diagnoses made before age 5, so findings may not apply to later-diagnosed children. The sample excluded children with cerebral palsy, neuromotor disorders, genetic or chromosomal abnormalities, and significant vision or hearing impairments. Analyses were limited to children with no more than mild cognitive or communicative impairments, so findings may not apply to children with moderate to severe intellectual disability. The sample included 15% female participants and most caregivers were white, so sex, gender and cultural diversity were limited. Anxiety was parent-reported and social communication differences were teacher-reported, which may reduce single-informant bias but may miss setting-specific patterns. The anxiety measure was developed for non-autistic children and may miss autistic-specific anxiety expressions. Executive function groups used general population norms, which may overestimate or oversimplify difficulties in autistic children. The behavioural regulation finding was a subgroup analysis, and ADHD symptoms were not measured. Only one year between assessments was captured, so longer developmental changes may differ.

The authors declared no potential conflicts of interest. The study was funded by Canadian Institutes for Health Research, Kids Brain Health Network, Autism Speaks, Government of British Columbia, Alberta Innovates Health Solutions and Sinneave Family Foundation. Some authors had fellowship or scholar support from Killam Trusts, University of British Columbia, Cordula and Gunther Paetzold Fellowship and Michael Smith Health Research BC.

Do not read the subgroup finding as evidence that anxiety improves social communication. It was an observational association only among children with clinically elevated behavioural regulation difficulties, and the overall sample showed no anxiety-social communication link.