Relationship Between Cognitive Abilities and ASD Severity over Time in Early-Diagnosed Preschoolers

Background: According to the most recent prevalence study conducted by the Centers for Disease Control, 38% of children with ASD have a co-occurring Intellectual Disability (CDC, 2008). Previous studies have found that children with average cognition or mild-moderate cognitive delays at the time of initial ASD diagnosis demonstrate increased cognitive scores at follow-up, whereas children with severe cognitive delays at initial diagnosis show stable scores over time (Flanagan et al., 2015). Limited research has been conducted examining the relationship between cognitive functioning and expression of ASD symptoms. Matson et al. (2008) found that IQ predicted the presence of ASD-like behaviors in adults without ASD, but IQ was not related to the expression of ASD symptoms in adults with the disorder. In contrast, developmental level was associated significantly with symptom severity, similarly, in toddlers with Autistic Disorder, PDD-NOS, and no ASD (Matson et al., 2012). However, Goldin et al. (2014) found that higher ASD severity scores were associated with lower developmental abilities in toddlers.

Objectives: The goals of this study were to 1) examine the stability and mean level change of cognitive abilities in children with ASD, and 2) determine the relationship between ASD symptom severity at initial diagnosis and later cognitive ability, in a sample of early-diagnosed children.

Methods: 35 children (86% male) were evaluated at two time points (T1 mean age=27.9 months; T2 mean age=52.3 months) using the ADOS-2, expert clinical judgment, and measures of cognitive functioning (T1: Mullen Scales of Early Learning; T2: Differential Abilities Scales-II). All children met diagnostic criteria at Time 1, whereas 30 children (85.7%) met criteria at Time 2. The sample was racially and ethnically diverse (57.1% White, 17.1% Black/African American, 8.6% Asian, 2.9% Arab, 8.6% multiracial; 17% Hispanic/Latino). ASD symptom severity was determined by total ADOS-2 algorithm score (T1) and ADOS-2 comparison score (T2).

Results: Children demonstrated significantly improved IQ scores between Time 1 (M=65.9, SD=13.7) and Time 2 (M=82.7, SD=35.1) (t(32)=-2.96, p<.01). A multiple linear regression model was calculated to predict overall IQ score at Time 2, based on overall IQ score and ASD symptom severity at Time 1. A significant regression equation was found (F(2,28)=11.65, p<.01) with R²=.45. Both overall IQ score and ASD symptom severity at Time 1 were significant predictors of overall IQ score at Time 2.

Conclusions: Children diagnosed with ASD at age 2 demonstrated significantly improved IQ scores at age 4, suggesting that early cognitive abilities may not be stable in children with ASD; once ASD symptoms are directly targeted in treatment, children may demonstrate increased capacity for learning. In addition, both overall IQ score and ASD symptom severity at age 2 uniquely predicted overall IQ score at age 4. This finding has implications for treatment planning for children diagnosed with ASD before 36 months, as all children in this sample received early intervention services prior to and between initial diagnosis and follow-up.