(PS2-61) Factors That Lead to the Identification of Autism in a School-Based Sample

Abstract Text:

Background: Autism identification in medical and educational settings can be challenging. Reasons that children with autistic traits might be missed or misdiagnosed include symptoms that overshadow this diagnosis (e.g., emotion dysregulation), or systemic barriers to receiving a medical evaluation (Golson et al., 2022). The aim of the study is to explore the above differences between children with and without preexisting autism identifications who were referred by school providers to a study for autistic students with anxiety.

Methods: The current sample (n=183) is derived from a school-based intervention study for autistic youth with anxiety. Students were referred by interdisciplinary school providers (e.g., educators, counselors, SLPs) following an initial training about autism and anxiety. Inclusion criteria: (1) an established autism identification/diagnosis and/or (2) meet clinical cutoff on the Social Communication Questionnaire, Social Responsiveness Scale, 2nd Edition, or Autism Diagnostic Observation Schedule – 2nd Edition (ADOS-2). Children who met criteria for the study were between the ages of 8 to 14 years old, had an IQ above 70, concerns for anxiety, and had an IEP or 504 plan. Average age of the students in this sample was M=10.30 (SD=1.68). Approximately 70% had been identified as autistic, 75% were male and 1% were gender diverse. Approximately 60% identified as White, 14% Black, 6% Asian, 3% American Indian, 12% mixed race, and 18% identified as Hispanic/Latino.

To address the study aim, measures of autism symptom severity (SRS-2), internalizing and externalizing symptoms (Strengths and Difficulties Questionnaire [SDQ], Emotion Dysregulation Inventory [EDI], Screen for Child Anxiety Related Disorders [SCARED]), and demographic information (parental income, parental education, race and ethnicity, and parental employment status) were collected. We hypothesized that children with lower socioeconomic status, minoritized racial and ethnic background, lower levels of autism characteristics, and higher levels of externalizing and internalizing concerns would be less likely to have formal autism identifications.

Results: Chi-square tests revealed that racial and ethnic background, household income, parent education, and gender were not associated with likelihood of having a formal autism diagnosis (p = 0.30- 0.57). One-way ANOVA tests with Bonferroni correction identified that children with a formal autism diagnosis were more likely to have higher scores on the SRS-2 (F(1, 144) = 4.26, p = 0.04) and parent-reported SDQ Prosocial Skills (F(1, 143) = 13.04, p < 0.001) than children without an established diagnosis. All other child factors that were investigated and noted above were not significant (p = 0.10-0.48).

Discussion: In this school-based sample of children with autism or suspected autism, children with lower levels of autism-related characteristics and greater prosocial skills were less likely to have obtained a formal autism diagnosis. Socioeconomic and demographic variables were not related to likelihood of having an established diagnosis. Within this sample, systemic barriers and diagnostic overshadowing did not explain differences between formal vs. informal autism identification.