A Randomized Trial of the Accuracy of Novel Telehealth Instruments for the Assessment of Autism in Toddlers.

The sample was recruited from a clinical waitlist after referral for concern about autism or developmental delay, and 92% were diagnosed with autism after in-person assessment. The study included only a small number of children without an autism diagnosis, so it cannot show how well tele-assessment works in less-selected community toddlers. Tele-assessment was completed in a controlled clinic room with research-provided equipment and materials, not in the home. All families had tele-assessment before in-person assessment, not in a counterbalanced order, so fatigue or prior exposure may have affected in-person scores. The remote STAT was a non-standardized adaptation, and ADOS-2 administration was modified for pandemic precautions for some participants. Clinical interviews were not standardized or coded across clinicians. The three children incorrectly identified as having autism were very young, and the study could not determine whether tele-assessment is less accurate for very young toddlers.

Do not conclude that tele-assessment can replace in-person evaluation or rule out autism. Eight children diagnosed with autism after in-person assessment were missed by tele-assessment, and the study included toddlers referred because of concern for autism or developmental delay.