Underlying Factor Structure of the Brief-2 in a Clinical Sample of Children and Adolescents with ASD

Background: Children and adolescents with autism spectrum disorder (ASD) often struggle with executive functioning (EF) skills, which can lead to social mistakes and challenges with making decisions, as well as problems with initiating, following, and organizing plans. Given that EF deficits have a significant impact on adaptive functioning of youth with ASD, the measurement of EF skills has become an important component of their neuropsychological assessments. The Behaviour Rating Inventory of Executive Function (BRIEF-2) is a standardized questionnaire that assesses the extent to which a child regulates and guides their behaviours in everyday environments (Gioia et al., 2015). The BRIEF-2 contains nine scales that form three indexes (Behavioural Regulation Index, Emotional Regulation Index, and Cognitive Regulation Index) and one composite score (Global Executive Composite). This is different than the previous version of the questionnaire, which contained eight scales across two scales: Behavioural Regulation Index and Metacognition Index. Despite the adoption of the second edition of the BRIEF within current ASD literature and in clinical settings, there has been no published research to date addressing the psychometric properties of the BRIEF-2 for youth with ASD.

Objectives: We investigated the underlying factor structure of the BRIEF-2 in a clinical sample of youth with ASD.

Methods: The BRIEF-2 parent form was completed by parents of 58 youth with ASD (52 males), ages 5 years to 16 years of age (M=9.64 years, SD=3.47). All participants had a community diagnosis of ASD, which was confirmed by parent report of ASD symptomatology on the Social Responsiveness Scale-2.

Results: BRIEF-2 scale scores (T-scores) were subjected to principal axis analysis (Direct Oblimin Rotation with Kaiser normalization). The rotated solution found two factors, based on eigenvalues greater than 1 rule and examination of the scree plot (Figure 1), which explained 49.39% of the variance with factor 1 contributing 39.54% and factor 2 contributing 9.86% (see Table 1). Factor 1 primarily contained variables related to cognitive regulation, whereas factor 2 included variables related to emotional and behavioural regulation. As well, self-monitoring (which should load onto behavioural regulation according to BRIEF-2) loaded on both factor 1 and 2.

Conclusions: Despite the fact that the BRIEF-2 manual reports a 3-factor structure for this measure in non-ASD youth, our results revealed a 2-factor structure instead. As such, evidence to support the validity of the three-factor structure of these constructs in youth with ASD was not found. Specifically, the results from our sample did not support distinct emotion regulation and behavioural regulation indices. In other words, our data did not support the underlying theoretical model as described in the BRIEF-2 manual. Therefore, conclusions about cognitive regulation, emotional regulation and behavioural regulation should be made cautiously and differential comparisons of these skills may not be warranted.