Using the childhood autism rating scale to diagnose autism spectrum disorders.

Chlebowski et al. (2010) tested how well the Childhood Autism Rating Scale (CARS) spots autism in toddlers and preschoolers.

They looked at 2- and 4-year-olds with autism, other delays, and typical development.

The team checked which CARS cut-off scores best separated the groups.

A CARS score near 25 caught most children with any autism spectrum disorder.

Scores near 30-32 helped tell classic autism from the milder PDD-NOS group.

The cut-offs worked well for both age sets in the study.

Norris et al. (2010) pooled data on several screens and agreed CARS is solid, while tools like GARS miss too many cases.

Peiris et al. (2022) later used the same CARS cut-offs in Sri Lanka and found a cheap seven-item checklist tracked CARS well, so the numbers hold in low-resource clinics.

Barnard-Brak et al. (2016) showed the SCQ screen lost accuracy as kids aged, reminding us that good cut-offs in one study can weaken later—so far CARS has stayed steady, but keep watching.

You can feel confident using a CARS cut-off of 25 to flag possible ASD and 30-32 to judge severity in toddlers and preschoolers.

If a family lacks access to long tests, pairing these thresholds with a brief behavior list, as Hashan et al. did, may still give reliable results.

Start your next young-child evaluation with these numbers, then back them up with direct observation and parent report.