(PS1-64) Emotional Dysregulation in Emerging Adult ADHD: A Key Consideration for Accurate Classification of Impairment and Co-occurring Internalizing Problems

ADHD in emerging adulthood (18-29 years) has been linked to functional impairment across academic, occupational, and social domains, as well as increased risk for depression and anxiety (Anastopoulos et al., 2016). Yet, prior work has suggested that ADHD symptoms, in isolation, may not be sufficient to explain ADHD’s relations with these clinically relevant issues (Goh et al., 2023). Such findings support the idea that consideration of non-symptom risk markers may be crucial in explaining ADHD’s link with impairment and internalizing problems, with implications for more comprehensive assessment and treatment of ADHD. One such risk marker is emotional dysregulation (ED), described as the expression and experience of emotions which are excessive relative to social norms, context, and developmental stage (Shaw et al., 2014). Studies have suggested rates of ED as high as 70% in adults with ADHD (Shaw et al., 2014), with both ADHD symptoms and ED uniquely associated with outcomes. As such, incorporating measurement of ED into ADHD assessment protocols could help facilitate unique insights into risk for specific types of impairment and inform transdiagnostic intervention aimed at issues that commonly co-occur with the disorder but are not necessarily encompassed by symptoms. Yet, there remains a lack of understanding in how to best incorporate ED into assessment protocols in a robust and data-driven manner. The current study thus sought to use machine learning techniques to clarify ED’s incremental validity, beyond ADHD symptoms, in classifying functional impairment and co-occurring internalizing problems during emerging adulthood, and harness it to develop an algorithm that could be easily deployed in time-limited settings to facilitate more comprehensive and transdiagnostic assessments and interventions.

Participants were 1,539 college students aged 18-25 years with diagnosed ADHD (M_age = 19.5 years, 30% male, 67% White/Non-Hispanic, 18% Hispanic/LatinX) who completed measures of ADHD symptoms, ED, functional impairment across multiple domains, and depression and anxiety. Comparison of a random forest model only including ADHD symptoms versus one with ADHD symptoms and ED dimensions suggested that the inclusion of ED significantly improved model performance across all external correlates (ps < .001). The ADHD + ED classification model also demonstrated acceptable to excellent performance in an independent sample (AUCs ranged from .70-.85). Variable importance analyses suggested inattention sum scores and the Limited Access to Emotional Regulation Strategies dimension as the most important features for model classification. Results provided support for ED as playing a key role in explaining ADHD’s relation with impairment and co-occurring internalizing problems during emerging adulthood, with machine learning providing a means to facilitate actuarial classification of these difficulties. Given the importance of the Limited Access to Emotion Regulation Strategies dimension, focusing intervention efforts on this dimension, in addition to inattention, could provide the most efficient and effective means of addressing impairment and co-occurring internalizing problems in emerging adults with ADHD.