Background: Data from epidemiology have consistently highlighted a disparity between the true prevalence of childhood psychiatric disorders and their recognition as defined by receiving a clinical diagnosis. Few studies have looked specifically at the level of unidentified autistic spectrum disorder (ASD) in the population.
Method: Logistic regression was used to determine the behavioural traits associated with receiving a diagnosis of ASD using data from the Avon Longitudinal Study of Parents and Children (ALSPAC). A composite score was derived to measure levels of autistic traits; undiagnosed children with scores matching those diagnosed with ASD were identified. Levels of educational provision beyond that provided by standard schooling were examined.
Results: Fifty-five percent of children with autistic traits at the same levels as those who had an autism diagnosis had not been identified as needing extra support from education or specialised health services. Of those who were identified as having special needs, 37.5% had been formally diagnosed with an ASD. For children with impairment at the same level as that associated with Asperger's syndrome, 57% had no special provision at school, and were not accessing specialised health services. Twenty-six percent of those who did have special provision at school had an ASD diagnosis.
Conclusions: The results suggest that there may be a substantial proportion of children on the autistic spectrum who are never identified by services.