Autism Screening in High-Risk Children in a Community Early Intervention Setting

Friday, May 12, 2017: 12:00 PM-1:40 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
D. Thao1, K. L. Reeb2, E. Hammer3, I. Toll3, M. Green3, S. Anderson3, M. Yudell4 and D. L. Robins5, (1)AJ Drexel Autism Institute, Philadelphia, PA, (2)AJ Drexel Autism Institute - Drexel University, Philadelphia, PA, (3)Elwyn SEEDS, Philadelphia, PA, (4)Drexel University School of Public Health, Philadelphia, PA, (5)Drexel University A.J. Drexel Autism Institute, Philadelphia, PA
Background:  Early detection of autism spectrum disorder (ASD) is imperative for access to ASD-specific intervention which leads to better outcomes for children with ASD. Children may be at increased risk for ASD due to a number of factors, including having an older sibling with ASD and/or being flagged for possible developmental delays. High-risk children are in need of timely referrals for ASD evaluation and ASD-specific intervention. While research suggests that the Modified Checklist for Autism in Toddlers (M-CHAT) is effective at screening in older and/or high risk children, there is limited research on the feasibility of screening within these populations.


  1. To assess three components of feasibility: demand, implementation, and integration of M-CHAT-R/F screening among intake staff at Elwyn SEEDS, which is the local education agency for preschool early intervention services (ages 3-5) in Philadelphia. 
  2. To measure preliminary validity of the M-CHAT-R/F in this population.


Elwyn SEEDS agreed to administer the M-CHAT, Revised, with Follow-Up (M-CHAT-R/F) during all intake calls. Intake staff attended training prior to deployment of the screening protocol. Archival records (n=792) of children who completed intake between May and August 2015 were extracted and reviewed. Mean age at intake was 41.36 months (SD=10.59, range 27.40-70.13 months). Variables extracted from records included demographic variables, M-CHAT-R item responses and total score, diagnosis, eligibility, referral source, and interventions received.

Results:  Feasibility of M-CHAT-R administration during intake was variable. Demand was low: less than half of the files included a completed M-CHAT-R. Implementation was poor: 75 M-CHAT-R forms contained one or more errors. Integration showed some change following deployment of the screening protocol, 23% of children who screened positive (n=15) were referred for ASD evaluation compared to 3% of children who screened negative M-CHAT-R (n=5) and 9% of those who did not have an M-CHAT-R screen (n=30). There were 90 (24.59%) positive M-CHAT-R screens in this sample, of whom 15 were diagnosed with ASD; an additional 48 were diagnosed with developmental delays, but it is unknown if they were referred for an ASD evaluation. Three additional cases diagnosed with ASD screened negative on the M-CHAT-R. Estimates of psychometric properties were 83% for sensitivity, 75% for specificity, and 24% for PPV, but must be interpreted with caution due to poor adherence to the screening protocol.

Conclusions:  Screening is important for detecting ASD in children, which leads to better outcomes for a child. Demand, use of the tool, was low in a high-risk community setting. Implementation, correctly administering the tool, was poor, and integration, or incorporation of the tool into current processes, showed some improvement. The low completion rate suggests barriers to implementing screening during intake at Elwyn SEEDS. However, when used, the M-CHAT-R is successful at detecting many of the cases diagnosed with ASD. Therefore, it is essential to improve uptake of screening procedures in community settings. Strategies to increase feasibility of autism screening include qualitative studies to determine the barriers to screening, completing evaluations, ongoing training for the staff administering the MCHAT-R and understanding early intervention providers’ beliefs.