Screening for Autism Spectrum Disorders in Underserved Populations: Global Versus Specific Surveillance Instruments

Abstract

Objective: The identification of developmental delays during well-child visits is insufficient with clinical judgmentalone (30-50%).Method: We analyzed the usefulness of clustering communication and social-personal areas from neurodevelop-mental screening ASQ-3 and Denver II PRUNAPE questionnaires, compared with M-CHAT and ADOS as the initialscreening for detection of Autism Spectrum Disorders (ASD) at well-child controls at primary-care settings in low-and middle-income countries (LMIC).Results: One hundred children at the Department of Pediatrics from 12 to 66 months (average: 35 months), wereassessed by a developmental pediatrician, trained and supervised by ECHO-model. The assessment was completedwith a full assessment for autism with a child neurologist, and a child-juvenile psychiatrist. Thirty-one of 100 chil-dren arrived at autism spectrum disorder diagnostic criteria.Conclusion: Failure at ASQ-3 communication plus social-personal clusters was in good agreement with M-CHATto predict risk criteria for autism compared with a full assessment protocol. Clustering Communication and Person-al-Social developmental skills from global screeners were sufficient at the pediatric visits, let differential diagnoseswith global developmental delay, and communication developmental disorders. These scales covered a wide range ofages for early identification of children with an autism spectrum disorder in primary care settings within an on-siteor online format