Online administration of the ADOS for research with adolescents and adults in response to the pandemic.

Only 37 participants took part, and they were verbal adolescents and adults who could use Module 4, so the results do not tell you about younger children, people with less speech, or people needing other ADOS modules. There was no in-person ADOS-2 comparison in the same people, so the study cannot show whether online scores match standard assessment. The online version omitted the Break activity, which is meant to show how a person responds when the examiner withdraws and engages in unstructured conversation. Some parent measures were missing, and the paper says this limits the sensitivity of the validity analyses. Several ADOS-2 items had lower agreement, especially rapport, speech abnormalities, asking for information, gestures, social overtures, empathy, insight, and responsibility. Nuanced gaze changes were harder to assess on video, and hand movements outside the camera view could not be checked against in-person scoring. The sample was mostly White and non-Latinx, so it is unclear how these results apply to other racial or ethnic groups. Some participants had completed an ADOS at least seven years earlier, and the paper does not say whether that affected the online results. The authors call the results preliminary and do not support replacing in-person assessment.

The authors declared no conflict of interest. The article is marked as supported by NIH extramural funding.

Do not read this as proof that online ADOS-2 can replace in-person diagnosis. The study had no in-person comparison, only tested verbal adolescents and adults on Module 4, and the authors state the results are preliminary and should be used as one component of a thorough evaluation.