Understanding Prevalence and Kindergarten Behavioural Profiles of Children with Autism Spectrum Disorder

Background: Health disorder such as Autism Spectrum Disorder (ASD) in early childhood can impact the developmental trajectory of the child. Early identification can facilitate access to information and resources that will assist the family and child to achieve the most optimal developmental outcomes. Identification of behavioural profiles common to children with ASD in preschool and kindergarten can provide helpful information for educational professionals in managing classrooms and promote inclusion. Building on our previous work investigating kindergarten behavioural profiles of children who were later diagnosed with Autism Spectrum Disorder, we have linked the data reported by teachers in kindergarten on child’s diagnosis and an assessment of child development in five domains with administrative data. In this study we are using the data from one Canadian province, Manitoba, to explore prevalence on ASD in kindergarten, teachers’ knowledge (determined by linking), and behaviour of those children.

Objectives: This study aims to 1) determine the concordance of EDI and administrative data when identifying children with ASD in Manitoba, and 2) examine differences in behavioural profiles of children in five developmental domains by child’s diagnosis and teacher’s knowledge. We expect that there will be no difference in teachers’ ratings of children with the confirmed (through linkage) ASD diagnosis between those for whom teacher reported the diagnosis and for whom s/he did not.

Methods: The Early Development Instrument (EDI) has been used in Manitoba since 2005 and has included the option to identify children that fall on the autism spectrum since 2011. Combining this population-level data with health and educational administrative data provides a powerful opportunity to explore the identification of ASD in Manitoba.

EDI data including indication of ASD were collected in Manitoba in academic years 2010-2011 and 2012-2013. Health and education administrative data have been collected on an ongoing basis in Manitoba and are housed through the Manitoba Centre for Health Policy (MCHP). These data include, but are not limited to, prescription information, hospitalization records, and codes for diagnoses.

Results: Results show good concordance between the administrative and EDI data. The administrative data show an overall prevalence rate of ASD in kindergarten at 0.9% (n=229) and the EDI data estimates ASD prevalence at 0.8% (n=191). Fifty-eight percent (n=177) of ASD cases were identified by both the administrative data and the EDI, 37% (n=112) of ASD cases were identified prior to kindergarten only by the administrative data, and 5% (n=14) of cases were identified only by the EDI data. Comparison of behavioural profiles between the groups is ongoing.

Conclusions: The EDI and administrative data in Manitoba show good concordance in identifying children with ASD at school entry. Our investigation takes into consideration details such as the age of first diagnosis and post-kindergarten diagnoses. We also examine associated child vulnerability at school entry as determined by the EDI, which has implications for practices, procedures, and policy surrounding identification and assistance of children with ASD in the school system.