Assessing Emotional/Behavior Problems in Children with ASD: Differences in ABC and CBCL Profiles By Language Level
There has been an increasing focus on assessment of social-communication, cognitive and language skills in minimally verbal (MV) children (Kasari et al., 2013), but little attention to psychiatric comorbidities in this clinical population. The Child Behavior Checklist is commonly used to assess emotional/behavior problems (EBP; Achenbach & Rescorla, 1991), but includes items reflecting internal experiences that may be difficult to evaluate in MV children. The Aberrant Behavior Checklist (Aman et al., 1985) was developed to assess EBP in individuals with intellectual disability (ID); it may be well-suited for assessing MV children but miss more internalized experiences of verbal children. Previous studies suggest the ABC is appropriate for use with children with ASD (Kaat et al., 2014); no study has examined whether item response patterns vary by child language level.
Examine the utility of the ABC and CBCL for assessing EBP in children with ASD across different language levels and use items from the ABC to gain insight into how EBP may manifest differently in MV vs. verbal children.
Participants were 1,964 6-18 year olds with ASD from the Simons Simplex Collection. ABC (Irritability, Lethargy, Hyperactivity) and CBCL (Internalizing, Externalizing) were used to examine EBP. ADOS module was used as a proxy for language level (Module 1=MV, 2=phrases (P), 3=verbally fluent (VF)). One-way Analysis of Variance with post-hoc Bonferroni tests were used to compare EBP across language levels. Chi square analyses were used to assess language group differences in a)children meeting clinical cut-offs for CBCL subscales and b)ABC item response patterns. Correlations between EBP, ASD symptoms and IQ were assessed within language groups.
Each ABC subscale showed language group differences (all p<.001). MV children had higher Lethargy scores than P and VF groups (p<.001), but differed from only the VF group on Hyperactivity and Irritability (MV>VF, p<.001). On the CBCL, language groups showed different Internalizing (p<.001), but not Externalizing scores (p=.48). In contrast to the ABC, the VF group had higher Internalizing scores than MV or PS groups (p<.001). Within the Internalizing domain, only 1% of the MV group fell in the range of clinical concern (t-score ≥ 70) on the Anxious/Depressed subscale, compared to 4% of the PS and 20% of the VF group (p<.001). Scores on other Internalizing subscales did not differ by language level.
Across the three ABC scales, seven items showed similar response patterns across language groups. Scales also showed differential relationships with IQ and ASD symptoms across language levels, including within the MV group. Language group differences in item distributions and associations with other child characteristics will be discussed with respect to differences in manifestation of ASD symptoms and overlap with EBP in MV vs. verbal children.
The ABC and CBCL showed different patterns of EBP across language levels, underscoring a need for careful consideration of the assessment of EBP in MV children. Findings have implications for the use of ABC and CBCL in large samples of varying language abilities, as EBP may manifest differently in MV and verbal children.