Differences in Parent Reported Adaptive and Executive Functioning Between African American and White Children with ASD

Background:  Research into cultural differences in the symptom profile of autism spectrum disorder (ASD) is limited. However, prior studies have reported that children from ethnic minority backgrounds display greater impairments in several areas, including social symptoms, language, cognition, and motor skills (Cuccaro et al., 2007; Tek & Landa, 2012). Prior studies have utilized early childhood samples, including children with developmental delays. Further research is needed to investigate ethnic differences in symptoms, particularly in older and higher-functioning samples.

Objectives: The goal of the present study was to examine differences between African American (AA) and White children with ASD in parent-reported adaptive behavior and executive function. Based on prior research, it was hypothesized that AA children would have greater impairments in both domains. 

Methods:  Participants included children with a confirmed diagnosis of ASD, ages 5-18 years with an IQ>70, recruited as part of a larger research study and through clinical evaluation. From the initial sample of 966 participants, the final sample of AA (n=38) and White (n=233) participants with complete data was generated. There were no significant differences between the groups on age, gender, diagnosis of ADHD, or ADHD medication use; however, AA participants had significantly lower IQ scores (t=3.28, p<.01). Parent report on the Behavior Rating Inventory of Executive Function (BRIEF; Gioia, Isquith, Guy, & Kenworthy, 2000) and the Vineland Adaptive Behavior Scales-II (Sparrow, Cichetti, & Balla, 2005) was compared across the AA and White groups. 

Results: On the BRIEF, White children showed significantly greater impairment on the Global Executive Composite (GEC; t=2.09, p<.05) and the Behavior Regulation Index (BRI; t=2.37, p<.05), and greater impairment approaching significance on the Metacognitive Index (MCI) of the BRIEF (t=1.84, p<.10). Hierarchical multiple linear regression analyses indicated that ethnicity was no longer a significant predictor of BRIEF scores after controlling for IQ, gender, ADHD diagnosis, and medication use. The results of stepwise regression analyses entering all variables into the model and statistically selecting the model that accounted for the greatest proportion of variance indicated that gender best predicted both GEC (F= 5.78, p<.01) and MCI (F=6.36, p<.05) scores, while a model using both gender and ethnicity best predicted BRI scores (F=4.06, p<.05).

 No significant group differences were observed in Vineland scores. Stepwise regression analyses indicated that IQ scores best predicted Vineland total scores in the overall sample (F=17.69, p<.001) and within the White sample (F=17.76, p<.001), with a small effect size (R²=.08, both samples). Within the AA sample, IQ did not significantly predict Vineland total scores, likely due to reduced power within the AA sample, but did significantly predict communication scores (F=8.56, p<.01). 

 Conclusions:  The results suggest a mixed profile of ethnic differences. Although ethnic differences were observed on the BRIEF, these were better accounted for by other factors. In contrast to prior studies, no differences were found on in adaptive behavior. The present study’s use of an older sample without intellectual disability may suggest that ethnic differences in prior research are best explained by higher rates of developmental delay in ethnic minority children with ASD.