Leveraging a Statewide Early Intervention System to Reduce Autism Evaluation Wait Times

ObjectiveTo evaluate the diagnostic accuracy of an Early Intervention (EI) autism diagnostic pathway, which included: (a) a multi-disciplinary team, (b) direct child observation and parent interview, and (c) a feedback session to explain the diagnosis.MethodsChildren between 15 and 36 months were referred to this prospective double masked diagnostic study. The primary outcome was agreement on autism diagnosis between the EI (index diagnosis) and the expert diagnostic team (reference diagnosis). One observational assessment (TELE-ASD-PEDS) and one caregiver interview (Toddler Autism Symptom Inventory) were used to inform the diagnosis. ResultsIn a sample of 639 children (mean age: 30 months; 70% male; 15% Hispanic, 30% Black, 32% White), we found 93.2% [95% CI, 91.0%-95.2%] agreement between EI teams and the expert team across conclusive diagnoses. Diagnostic accuracy indices related to a positive autism diagnosis (i.e., specificity, PPV) were higher (96.7% [95% CI, 95.0%-98.2] and 95.6% [95% CI, 93.7%-97.2%] respectively) than indices related to a negative autism diagnosis (i.e., sensitivity [67.2%; 95% CI, 56.1%-78.2%], NPV [73.8%; 95% CI 61.9%-84.1%]). The overall rate of autism diagnosis was 80%. Parents also reported high satisfaction. ConclusionEI teams are very accurate at diagnosing autism in toddlers enrolled in EI without any concomitant medical conditions. In the current study, 80% of children could be triaged to an EI diagnostic pathway, such that the limited number of experts could be reserved for more complex or ambiguous cases. These findings suggest that this pathway has the potential to markedly reduce delays in autism identification