Can a broadband developmental-behavioral screening test identify children likely to have autism spectrum disorder?

Clin Pediatr (Phila). 2007 Nov;46(9):801-5. doi: 10.1177/0009922807303928. Epub 2007 Jul 19.


Because services for children with autism spectrum disorder (ASD) are scarce, when children fail a broadband screening measure, providers need to carefully discern which children need ASD evaluations and which do not. This research considers how well a broadband screening test sorts those with and without probable ASD. The subjects were 427 children between 18 and 59 months of age with elevated risk scores on broadband screening, ie, Parents' Evaluation of Developmental Status (PEDS), a 10-item measure eliciting parents' concerns. Parents also completed the Modified Checklist of Autism in Toddlers (M-CHAT), an autism specific screen. The results showed that of the 427 children at risk on PEDS, 34% (N = 144) passed the M-CHAT. To determine whether these potential overreferrals could be reduced, parents' concerns on PEDS were used to predict M-CHAT results. Three or more discrete types of concerns, varying by age, characterized children who failed the M-CHAT while fewer than 3 were associated with passing. This reduced overreferrals by 70% while maintaining high levels of sensitivity (81%). Although compliance with the American Academy of Pediatrics recommendations for both broadband and autism-specific screening at 18 and 24 months is still recommended, viewing performance patterns on a broadband screening test can substantially reduce overreferrals to autism specialty services.

MeSH terms

  • Autistic Disorder / diagnosis*
  • Child Behavior / classification*
  • Child, Preschool
  • Humans
  • Infant
  • Infant Behavior / classification*
  • Logistic Models
  • Parents / psychology*
  • Surveys and Questionnaires