New autism diagnostic interview-revised algorithms for toddlers and young preschoolers from 12 to 47 months of age.
Switch to the toddler ADI-R algorithm for kids 12-47 months to cut missed diagnoses.
01Research in Context
What this study did
The team built new scoring rules for the Autism Diagnostic Interview-Revised.
They wanted a version that works for toddlers and young preschoolers aged 12 to 47 months.
Kids with autism, typical kids, and kids with other delays all took part.
What they found
The new toddler rules caught more autism cases and made fewer false alarms.
Sensitivity and specificity both improved compared with the old algorithm.
How this fits with other research
Lancioni et al. (2006) warned that the original ADI-R often missed toddlers. They saw that many little ones had not yet shown repetitive behaviors.
Kim et al. (2012) fixed the gap. They lowered the behavior count needed and added age-based cut-offs.
Saemundsen et al. (2003) showed ADI-R and CARS only agreed about two-thirds of the time. The toddler update narrows that gap by making the interview more sensitive early on.
Hedley et al. (2015) found the 10-minute ADEC play screen also caught most ASD toddlers. The new ADI-R rules give you a second, deeper tool when you need a full diagnosis.
Why it matters
If you assess toddlers, ask your team which ADI-R algorithm they use. The toddler version gives cleaner answers and saves families from repeat visits. Swap the disk or update the software, then trust the lower cut-offs for kids under four.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Email your assessment team to confirm the toddler ADI-R algorithm is installed and select the 12-47 month option before your next intake.
02At a glance
03Original abstract
Autism Diagnostic Interview-Revised (Rutter et al. in Autism diagnostic interview-revised. Western Psychological Services, Los Angeles, 2003) diagnostic algorithms specific to toddlers and young preschoolers were created using 829 assessments of children aged from 12 to 47 months with ASD, nonspectrum disorders, and typical development. The participants were divided into three more homogeneous groups by language level and age. Items that best differentiated the diagnostic groups were selected and arranged into domains based on multifactor item-response analyses. Using the new algorithms for toddlers and preschool children, we were able to improve sensitivity and specificity compared to the previously developed algorithm.
Journal of autism and developmental disorders, 2012 · doi:10.1007/s10803-011-1213-1