Development of a New Measure to Identify Fundamental Emotion Regulation Processes Across the ASD Spectrum: Application to the Autism Inpatient Collection

Background:  There is increasing recognition that aberrant emotion regulation is a critical process in autism spectrum disorder (ASD) that may underlie problem behaviors and psychiatric comorbidity (Mazefsky et al., 2013). The lack of emotion dysregulation measures that are appropriate for this unique population is a barrier to advances in this area.

Objectives:  The overall goal is to develop a measure of emotion dysregulation, the Emotion Dysregulation Inventory (EDI), which captures the full range of presenting concerns in ASD and is free of verbal loading. Specific aims included: 1) Evaluate the EDI’s psychometric properties, including potential verbal ability or nonverbal IQ (NVIQ) biases; and 2) Investigate emotion dysregulation patterns in psychiatric inpatients with ASD.

Methods: The Emotion Dysregulation Inventory (EDI) was developed based on guidelines from the Patient Reported Outcomes Measurement Information System (PROMIS), a NIH-initiative focused on creating measures of health status using modern psychometric methods (see Figure 1). Participants included 88 children (M=12 years) with ADOS-2-confirmed ASD who were admitted to an ASD specialized psychiatric inpatient unit as part of the Autism Inpatient Collection (AIC). The EDI and Aberrant Behavior Checklist (ABC) were completed by a consistent parent/guardian at admission and discharge. Participants had NVIQ standard scores ranging from 30-135 (M=71), and 57% (n=48) have limited verbal skills (ADOS-2, Module 1 or 2). Analyses focused on the total EDI intensity score. The EDI also includes past behavior time course ratings to aid in clinical conceptualization which will be explored in future analyses. 

Results: The EDI total score covered a wide range (45-229; M=125; SD=38) and was normally distributed. The EDI was correlated (r=.55) with the ABC Irritability Scale (p < .000), suggesting adequate convergent validity without excessive overlap. The EDI was not significantly correlated with age (r=-.06) or NVIQ (r=.07). EDI mean scores did not differ between minimally-verbal (M=121, SD=38) and verbal participants (M=128, SD=37), or between participants with NVIQs of ≤70 (M=124, SD=40) and NVIQs >70 (M=127, SD=36), p>.05. Internal consistency was high (Cronbach’s alpha =.95). Discharge EDI scores (M=67) were significantly lower than admission scores (M=125), p<.000. Variability in item-total correlations suggested multidimensionality, though the majority of items loaded strongly on the total score. Items with high item-total correlations captured increased negative affect, emotional reactivity, lability, and other indicators of poor emotion regulation, whereas items that typically indicate comorbid depression and anxiety had low item-total correlations. Exploration of item means suggested that emotion dysregulation in psychiatric inpatients with ASD is characterized by problems with sustained and rapidly escalating emotional reactions, with less frequent endorsement of symptoms of decreased vitality, decreased positive affect, and anxiety.  

Conclusions: Initial evaluation of the EDI suggests it will be an informative, change-sensitive measure of important emotional processes (emotional reactivity and poor regulation) in ASD that can be applied across the spectrum of verbal and IQ abilities. The EDI development process (e.g., PROMIS framework, pilot collection with a sample encompassing all verbal and IQ abilities) provides a model that can be applied to address the dearth of sensitive and valid measures for ASD.