Collaborating with Community and Health Care Service Providers: A Community-Based Screening Program for Identifying Toddlers with Autism Spectrum Disorder

Background:  Although research has shown that we can identify toddlers with autism spectrum disorder (ASD) in their first two years of life, the mean age of diagnosis in (non-research) clinical context is still around 4 years of age (CDC, 2012). This limits the access to interventions that are most effective when offered before 3 years of age. Knowing that developmental surveillance and targeted screening for ASD could help decrease the mean age at diagnosis, collaboration between community resources and health services should be supported to implement screening program for young children with ASD.

Objectives:  To explore the feasibility of a community-based screening program for toddlers (12-30 months old) with ASD and identify contextual components that facilitates or impedes its implementation.

Methods:  This prospective collaborative study uses a 3-step procedure to screen toddlers for ASD in the Eastern Townships (Quebec, Canada). Step 1: Parents first completed a one page screening questionnaire depending on the child’s age. Step 2: A 30-minute follow-up interview was completed with parents of children who scored at risk of ASD after the first step. Step 3: The research team completed the ADOS-T and the ADI-R with families still considered at risk after the follow-up. Descriptive analyses and predictive value was calculated separately for children aged between 12 and 18 months (group 1) who were screened via the Infant Toddler Checklist (ITC) and children aged between 18 and 30 months (group 2) who were screened using the Modified Checklist for Autism in Toddler – Revised (M-CHAT-R).

Results: 690 questionnaires were collected at step 1. Most (67%) of the questionnaires came from the community services or daycare (21%). Of the 690 questionnaires completed, 8.4% (n=58) scored at risk of ASD at step 1, 2.7% (n=15) remained positive at step 2, and 0.44% (n=3) finally scored positive at the third step. Thus, in this study, the positive predictive value of the ITC (group 1) was 0% and 23.07% for the M-CHAT-R (group 2). Information from the clinical setting revealed an increase in children of 3 years old and younger referred to the evaluation clinic for suspicion of ASD in the Eastern Townships between the beginning of the project (8% in 2014) and the time of analysis (33% in 2016).

Conclusions: Based on the result of the pilot project, the community-based screening program should be implemented in the Eastern Townships to continue identifying toddlers at risk of ASD. However, based on the results and recommendations of the AAP Bright Futures guidelines (2016) we proposed to target specifically older toddlers (i.e. 16-30 months) for which the M-CHAT-R can be used. It is important to highlight the collaborations with partners, specifically community services and daycare that were involved in the project. A short training on early markers of ASD might be useful so that they can target children at higher risk and encourage their parents to complete the screening questionnaire.