School Exclusion as a Partial Mediator Between Autism and Adolescent Mental Health

This study investigated whether school exclusion during secondary school mediates the association between Autism Spectrum Conditions (ASC) and adolescent internalising and externalising mental health difficulties.

We analysed data from the UK Millennium Cohort Study (N = 10,228, waves 1-7). Parents reported ASC diagnosis at ages 5-11 and whether their child had been excluded at ages 11-14. Internalising and externalising symptoms were self-reported by adolescents at age 17 on the Strengths and Difficulties Questionnaire. Confounders were mapped onto a Directed Acyclic Graph. Causal mediation analyses, using the g-formula for randomised interventional analogues, estimated direct and indirect effects while accounting for these confounders without biasing the causal pathway.

Autism was associated with higher internalising and externalising scores, with autistic adolescents more likely to experience school exclusion than peers. School exclusion partially mediated the association between ASC and externalising symptoms (indirect effect = 0.03 (95% CI: 0.004, 0.06, p = 0.024) . No evidence of mediation was found for internalising symptoms.

School exclusion acts as one modifiable environmental mechanism contributing to externalising difficulties in autistic adolescents, likely through the loss of protective school supports. Preventative, autism-informed strategies that reduce reliance on exclusion and provide appropriate support could form part of wider efforts to improve mental health outcomes in autistic adolescents.