The Association between Parent-Reported Emotional/Behavioral Problems and the Modified Checklist for Autism in Toddlers

Background: Children with Autism Spectrum Disorders (ASD) have difficulties in the areas of social-communication and other restricted and repetitive behaviors. The assessment of these core symptoms may be complicated by the presence of other co-occurring emotional and behavior problems (EBPs), which can make differential diagnosis challenging. EBPs may include externalizing symptoms such as overactivity and aggression, as well as internalizing symptoms such as mood problems and anxiety. Studies have shown the presence of EBPs decrease the specificity of ASD screening and diagnostic instruments.

Objectives: This study assessed the impact of parent-reported EBPs on an often used autism screener, the Modified Checklist for Autism in Toddlers, Revised (MCHAT R/f). In addition, the study examined the potential benefit of combining EBP data with the MCHAT R/f data to improve screening accuracy.

Methods: Participants included 473 children, 1.5-5 years of age, referred to one of three sites for ASD-related concerns. EBPs were assessed via the Child Behavior Checklist (CBCL) and autism symptoms were screened with the parent-report MCHAT R/f in a subset (N = 290). Children were also administered a full battery of diagnostic measures including the ADOS-2, with a final diagnosis assigned by a senior clinician.

Results: Overall, participants who were diagnosed with ASD had significantly lower externalizing EBPs compared to those who were not diagnosed with ASD (i.e., CBCL T scores - Externalizing [62.6 vs. 68.1]). Internalizing symptoms were not significantly different. Higher MCHAT-R/f scores (e.g., more ASD symptoms) and fewer CBCL- Externalizing symptoms were significantly predictive of an ASD diagnosis. MCHAT-R/f specificity was reduced if CBCL- Externalizing behaviors were present (i.e., from 44% to 27%). In addition, MCHAT-R/f Positive Predictive Value (PPV) was reduced from 86% to 71% if externalizing behaviors were present.

Conclusions: For toddlers referred for an ASD evaluation, parental report of significant EBPs, and in particular externalizing problems, often result in higher scores on an autism screening measure that may not be directly related to core autism symptoms (i.e., a false positive). Therefore, a combination of measures that assess EBPs and autism core symptoms will improve screening in this population. More specifically, results suggest that combining EBP findings from the CBCL with those of the M-CHAT R/f may improve the specificity of this instrument.