(PS2-C55) Emotion Reactivity and Regulation and Prospective Prediction of Non-suicidal Self-injury in an Adolescent Psychiatric Inpatient Sample

Over the past two decades, rates of self-injurious thoughts and behaviors among adolescents have increased. Previous studies indicate that non-suicidal self-injury (NSSI) is a stronger predictor of future suicidal behavior than is a past history of suicide attempts. Individuals with a history of NSSI report that this behavior functions to regulate emotions. Thus, it is theorized that emotion dysregulation and heightened emotion reactivity may increase risk for NSSI. However, little is known about what facets of emotion regulation and reactivity serve as vulnerabilities for NSSI. In addition, much of what is known about emotion regulation, reactivity, and NSSI is derived from cross-sectional studies; few prospective studies have investigated how emotion regulation and reactivity influence NSSI over time. Such information is important insofar as it may provide more specific candidates for targeted intervention. In the present study, adolescents ages 12–17 admitted to a psychiatric inpatient unit (N = 180; M_age = 14.89; SD_age = 1.35; 71.67% assigned female at birth) completed self-report measures of emotion regulation and reactivity during their index hospitalization. Past-month NSSI frequency (i.e., days of NSSI engagement) was assessed during hospitalization and six months post-discharge. Drawing on several established theories of NSSI (Four Function Model, Emotional Cascade Model, Experiential Avoidance Model), we hypothesize that the inability to access emotion regulation strategies, poor impulse control when distressed, difficulty engaging in goal-directed behavior, and all facets of emotion reactivity (i.e., sensitivity, persistence, and intensity) would emerge as predictive of NSSI six months later. A negative binomial regression model was used and, after controlling for age, sex, and NSSI history, three dimensions of emotion regulation - reduced goal-directed behavior (IRR = .88 [95% CI = .77 – 1.00], z = -1.99, p = .047), poor impulse control when distressed (IRR = .90 [95% CI = .81 – 1.00], z = -2.03 p = .043), and the inability to access emotion regulation strategies (IRR = 1.20 [95% CI = 1.06 – 1.35], z = 2.96, p = .003) - significantly predicted NSSI six months later. Our findings that limited access to emotion regulation strategies, poor impulse control while distressed, and reduced goal-directed behavior are in line with extant literature suggesting NSSI serves an emotion-regulatory function. Emotional clarity and awareness, two constructs in need of further study and definition, did not predict NSSI in this sample. Contrary to our hypotheses, no aspect of emotion reactivity predicted NSSI. This contradicts a biosocial theory positing that emotionally invalidating environments and a biological predisposition to heightened emotion reactivity interact to potentiate NSSI. Taken together, these findings maintain that emotion regulation should be a primary target of NSSI interventions and treatment. Future research should expand on identifying the mechanisms that influence the course of NSSI, particularly in adolescent inpatient samples, which are at risk for self-injurious and suicide-related outcomes. Doing so may effectively synthesize these extant models and give rise to a unified theory of NSSI.