TELE-ASD-PEDS

The TELE-ASD-PEDS is a novel tool developed specifically for remote assessment of young children. It is derived from the application of machine learning technology to a database containing gold standard diagnostic tool data from several hundred individuals with and without ASD (Adiani et al. 2019). The most predictive items for young children were derived and then applied to interactive, parent-led tasks with guidance from expert clinicians (i.e. 2 ADOS-2 trainers and 5 ADOS-2 research reliable psychologists). The TELE-ASD-PEDS includes 12 discrete social activities or social bids, including opportunities for free play, opportunities for physical play routines (e.g., peekaboo, tickling), and activities (e.g., bubbles, snack) that may prompt a child to request. Following administration, clinicians rate child behavior on seven items (e.g., socially directed speech and sounds, use of eye contact, atypical vocalizations, unusual or repetitive play), allowing for dichotomous (yes/no) as well as Likert (3 = behaviors characteristic of ASD clearly present; 2 = possible atypical behavior; 1 = behaviors characteristic of ASD not present) scoring regarding presence and severity of predictive symptoms. Specific behaviors coded and associated Likert-style ASD-PEDS scores are presented in Table ​Table33.