Measuring Difficulties in Emotional Regulation Among Young Adults with History of Homelessness: Psychometric Properties of the Ders-18

Background: Many young adults with histories of homelessness report the experience of traumatic events both before and after entering homelessness that have been linked to substance use and mental health issues, as well as difficulties with emotional regulation. Even though emotional regulation skills are vitally important for enabling youth to navigate the complex and chaotic settings typically associated with homelessness, there is a dearth of literature on emotional regulation within this population. The current study seeks to examine the relationship between emotional regulation and mental health among young adults who have experienced homelessness. Aims include assessment of: 1) construct validity of the short-form of the Difficulties in Emotional Regulation Scale (DERS-18) in a sample of formerly and currently homeless young adults, 2) concurrent validity of the DERS-18 with clinical scores of anxiety, depression, and PTSD, and 3) predictive validity of the DERS-18 through use of cell phone-based momentary measures of affect over the course of one week.

Methods: Data comes from a study of HIV risk among young adults who have experienced homelessness, including both those currently experiencing homelessness (n=99) and in permanent housing (n=120), collected between 2017 and 2019 in Los Angeles, CA. The study included baseline survey data collection and seven days of cell phone-based ecological momentary assessment surveys, prompted within random 2-hour intervals during waking hours. Structural Equation Modeling was used to complete Confirmatory Factor Analysis (for construct validity) and Path Models focused on the relationship between mental health symptomatology and emotional regulation (for concurrent validity). Linear regression models tested the link between affective intensity and DERS-18 through person-level means of positive and negative affect and affective instability via the Mean Squared Successive Difference (MSSD) (for predictive validity).

Results: The DERS-18 was found to have construct, concurrent, and predictive validity in our sample. Based on model fit indices, evidence for a five-factor solution of the DERS-18, excluding the Awareness subscale, was found. Clinical scores of anxiety, depression, and PTSD were found to have a significant, positive correlation with DERS, indicating increased symptomatology was associated with worse outcomes in emotional regulation. Both intensity of negative affect and instability of positive affect were associated with greater difficulty in emotional regulation, while increased intensity of positive affect was associated with decreased emotional regulation score, indicating increased skills in emotional regulation.

Conclusions: This study addressed several gaps in both homelessness and emotional regulation literature. Primarily, there is evidence that a shorter DERS-18 with five subscales can effectively be used to better understand skills in emotional regulation with populations that experience complex trauma, such as young adults who have experienced homelessness. Additionally, we found evidence for use of the DERS-18 with young adults with homelessness histories in the context of current, literal homelessness, as well as those residing in supportive housing contexts. Future work should consider the use of the DERS-18 to assess skills in emotional regulation, understand outcomes, and better tailor interventions for young adults who have experienced homelessness.