Parent-Reported Social Interaction and Race, but Not Observed Social Impairments, Predict Intervention Enrollment in Toddlers Diagnosed with ASD

Background: It is widely accepted that early intervention following early diagnosis leads to the best prognosis for children with autism spectrum disorder (ASD; MacDonald, Parry-Cruwys, Dupere, & Ahearn, 2014; Orinstein et al., 2014). However, some children who are diagnosed as early as age 2 do not enroll in intervention targeting their ASD symptoms, despite this recommendation from clinicians. It is unclear what factors may contribute to a parent’s ability or decision to enroll their child in intervention. It is possible that race, socioeconomic status, severity of their child’s symptoms, the parent’s perception of the child’s difficulties, or other factors may contribute to enrollment (Thomas et al., 2007). It is important to identify factors contributing to intervention access in order to increase participation in relevant services.

Objectives: To identify predictors of enrollment in ASD-specific intervention following initial ASD diagnosis at age 2.

Methods: Toddlers (n=110) were diagnosed with ASD at around age 2 after screening positive on the Modified Checklist for Autism in Toddlers (-Revised) (M-CHAT(-R); Robins et al., 1999, 2009) and Follow-Up, or being flagged by their pediatrician for possible ASD. Evaluations included ADOS, Vineland Adaptive Behavior Scales (-II; VABS), Mullen Scales of Early Learning (MSEL), and a developmental history. Information regarding enrollment in intervention was collected at a follow-up evaluation at age 4.

Results:  Of the 110 toddlers diagnosed with ASD at age 2, 71 (64.5%) enrolled in ASD-specific intervention whereas 39 (35.5%) did not; all but 8 toddlers participated in some type of early intervention (e.g., speech therapy). Logistic regression revealed that VABS(-II) Social Domain was predictive of enrollment, OR=.91, p<.001, whereas the ADOS Social Affect Calibrated Severity Score (ADOS SA CSS) was not predictive, OR=1.05, p=.68. Race was also predictive, as Black families (n=18) had 91% lower odds of enrolling in ASD-specific intervention than White families, OR=.09, p<.001, and other races (Hispanic n=8, Asian n=5, Multiracial n=8) had 82% lower odds of enrolling compared to White families, OR=.18, p=.001. Furthermore, in a multivariable model, VABS(-II) Social Domain scores, OR=.92, p=.001, and race (Black OR=.13, p=.001, other OR=.20, p=.004), independently predicted seeking intervention. Maternal education and child sex, age, cognitive, language, and symptom domain variables were not predictive, ps>.05.

Conclusions: Although all children received an ASD diagnosis at age 2, 35.5% did not enroll in ASD-specific intervention as recommended. Parent-reported social interaction (VABS Social Domain) was predictive of enrollment, but directly-observed social symptom severity (ADOS SA CSS) was not predictive. Therefore, parents who were more aware of, and possibly more concerned about, their children’s social functioning were more likely to seek ASD-specific intervention regardless of actual symptom severity. Furthermore, racial minority families were much less likely to enroll in intervention than White families. This has clinical implications for working with parents to address barriers to enrollment in intervention. Future research should examine barriers, such as cultural beliefs, mismatch of race between the family and intervention provider, and stressors on the family related to socioeconomic or racial status. Research should also work to identify strategies to overcome these barriers.