Factor Analysis of the Childhood Autism Rating Scale in a Sample of Two Year Olds with an Autism Spectrum Disorder.

All children had DSM-IV-TR diagnoses of Autistic Disorder or PDD-NOS, not DSM-5 ASD; another toddler study found that 29% of children with those DSM-IV-TR diagnoses no longer met criteria when DSM-5 criteria were applied. The sample was narrow in age (16 to 32 months), so the factor structure cannot be assumed for older children. CARS ratings were based partly on ADOS observation, so correlations with ADOS severity scores are not an independent validation. The Visual Response item did not load on any factor, and its wording may be interpreted differently by raters. The emotional reactivity and stereotyped behaviors and sensory sensitivities domains fit together less well than the social communication domain. The sample was mostly male (77.3%) and mostly White (73.0%), and 36 children with low mental age were excluded from ADOS analyses, so findings may not generalize to all toddlers. This was an exploratory factor analysis with an internal split-half check, not a confirmatory factor analysis in an independent DSM-5 sample.

Deborah Fein and Marianne Barton are co-owners of M-CHAT, LLC, which receives royalties from companies that incorporate the M-CHAT into commercial products. The data reported here came from the freely available paper version of the M-CHAT. Emily Moulton and Kathryn Bradbury declared no conflict of interest.

Do not conclude that the CARS is a validated, stand-alone diagnostic measure for DSM-5 ASD in toddlers. The sample used DSM-IV-TR diagnoses, CARS ratings were partly based on ADOS observations, and one Visual Response item did not fit any domain.