Predictors of outcome among high functioning children with autism and Asperger syndrome.

Szatmari P, Bryson SE, Boyle MH, Streiner DL, Duku E.

Journal of Child Psychology and Psychiatry 44 (4): 520-528. 2003

Bottom Line

Do nonverbal and language ability in children with high functioning autism (HFA) and Asperger syndrome predict how well children will do two and six years after diagnosis?

  • 68 children with an IQ of 68 and above were followed over time.
  • Tests were done of language skills and social communications skills when the group of children was first formed and were performed again two years and then six years later.
  • The association between language skills and outcome was greater in the participants with autism than in those with Asperger syndrome.
  • The Vineland Adaptive Behaviour Scales (VABS) was more predictive of later outcome than were tests of grammar or vocabulary, likely because they lack a social component.
  • Working on language skills may be more likely to improve the outcome in children with autism than it would in children with Asperger Syndrome.

Question: Do measures of nonverbal and language ability in high functioning children with autism and Asperger syndrome independently predict outcome in some or all domains of functioning approximately two and six years later?

Background: People who have been diagnosed as having autism or Asperger syndrome retain their diagnostic features into adulthood, although there can be either a lessening or worsening of their symptoms.

Design: Time series study (subjects are followed over time and are periodically tested).

Setting: South central Ontario, Canada

Participants: Sixty-eight children with a formal diagnosis of autism or Asperger syndrome, with an I.Q. on the Leiter of 68 or above or 70 or above on the Stanford-Binet, were included in this group of children followed from the time of their first assessment or their entry into a treatment program (an inception cohort).

Assessment of prognostic factors: Language and social communication skills were tested when the cohort was formed. The scales used were repeated at intervals to assess what changes had taken place over time.

Main outcome measures: Autism Behaviour Checklist, the Vineland Adaptive Behaviour Scales (VABS) and several language scales.

Main results: The association between language skills and outcome was greater in the participants with autism than in those with Asperger Syndrome. Parental report from the Vineland Adaptive Behaviour Scales on communication skills strongly predicted outcome.

Conclusions: It appears from this study that the factors that determine outcome in children with autism are different than in children with Asperger syndrome.

In particular, language skills in children with autism are better at predicting their outcome. One test, the VABS appears to be a more accurate measure of "real-world" use of communication skills. It was more predictive of later outcome than were tests of grammar or vocabulary, likely because they lack a social component. Working on language skills may be more likely to improve the outcome in children with autism than it would in children with Asperger Syndrome.

Findings in children with autism cannot be used to determine outcome in children with Asperger Syndrome or other PDDs. This is especially true of nonverbal skills, which can help to predict outcome for autistic symptoms, particularly in children with Asperger Syndrome, but to a lesser degree amongst children with other PDDs. Overall, this study demonstrated that there is a need to test out how crucial language skills are to outcome in autism through well-designed randomized controlled trials.