Replication study of ADOS‐2 Toddler Module cut‐off scores for autism spectrum disorder classification. Issue 6 (8th March 2021)

• Record Type: Journal Article
• Authors: Hong, Ji Su; Singh, Vini; Kalb, Luke; Ashkar, Ahlam; Landa, Rebecca
• Is Part Of: Autism research. Volume 14:Issue 6(2021)
• Page Start: 1284
• Page End: 1295
• Publication Date: 2021-03-08
• Subjects: autism spectrum disorder -- classification -- diagnosis -- validity; Autism -- Periodicals; Autism -- Research -- Periodicals; 616.85882005
• DOI: 10.1002/aur.2496
• Languages: English
• ISSNs: 1939-3792

Abstract

Abstract: The Autism Diagnostic Observation Schedule, Second Edition (ADOS‐2) has been regarded as the gold standard assessment of autism spectrum disorder (ASD). While clinical validity of ADOS‐2 Modules 1–4 have been extensively studied, there has been very limited research examining the clinical validity of ADOS‐2 Toddler Module. The goal of this study was to examine alignment of the ADOS‐2 Toddler Module classification with clinicians' determination of ASD, with assessing cut‐off scores for diagnosing ASD in toddlers. A total of 412 toddlers ages 12–30 months who received ADOS‐2 Toddler Module as well as a best estimate clinical (BEC) diagnosis, were included in this study. ADOS‐2 Toddler Module cut‐off scores were determined using the appropriate algorithms (Algorithm 1 for toddlers aged 12–20 months and those aged 21–30 months with <5 words, Algorithm 2 for toddlers aged 21–30 months with 5 words or more). Receiver operating characteristic (ROC) curves were used to assess cut‐off scores that optimized sensitivity and specificity when compared against BEC diagnosis. The selected cut‐off scores were examined using frequency tables to compare ADOS‐2 classification against BEC diagnosis. For Algorithm 1, classification statistics were optimized at the cut‐off score of 12 with an area under the curve (AUC) of 0.92. For Algorithm 2, classification statistics were optimized at the cut‐off score of 10 with an AUC of 0.96. The ADOS‐2 Toddler Module classification is strongly aligned with BEC diagnosis. The optimal cut‐off scores identified in the current study reflect the same results configured by the prior study. Lay Summary: ADOS‐2 Toddler Module has been widely used for the ASD assessment, but there have been limited research on its clinical validity. This study is the first replication of the ADOS‐2 Toddler Module with a large independent sample. We examined alignment of the ADOS‐2 Toddler Module classification with clinicians' determination of ASD, with assessing cut‐off scores, and confirmed the clinical validity of ADOS‐2 Toddler Module. Cut‐off scores of ADOS‐2 Toddler Module cited in the manual yielded best clinical utility for diagnosing ASD in toddlers.