A Longitudinal Examination of the Association Between Non-Suicidal Self-Injury, Emotional Intelligence, and Family Context in Adolescents

Alexandra Perloe

Advisor: Christianne Esposito-Smythers, PhD, Department of Psychology

Committee Members: Timothy W. Curby, Tara Chaplin

Research Hall, #161
July 20, 2015, 12:00 PM to 09:00 AM


Non-suicidal self-injury (NSSI), which peaks in adolescence, is associated with poor psychological health, physical pain, and risk for future suicide attempts, but empirically supported treatment is scarce. Emotional intelligence (EI) – the ability to perceive, understand and manage emotions –captures several factors relevant to the development of NSSI. The present set of studies examined the role of EI in the development of NSSI in a clinical adolescent sample. The first study examined the potential for a reciprocal longitudinal relationship between NSSI and four facets of EI (stress management, intrapersonal, interpersonal, and adaptability) that have theoretical and empirical links to NSSI. It was hypothesized that poor EI would increase the likelihood and frequency of future NSSI, which would in turn prevent adolescents from learning more adaptive coping skills and thus exacerbate their EI deficits. Participants were 91 adolescents ages 13-18 (mean age= 15.5, SD = 1.4; 61% female; 46.1% Caucasian) who were enrolled (along with one parent each) in a randomized controlled trial of a prevention workshop targeting suicide, HIV and substance abuse. All adolescents had received mental health treatment at baseline. Results of cross-lagged autoregressive models did not support a bidirectional relationship between NSSI and EI over time, suggesting that cross-sectional correlates of NSSI may not hold longitudinal predictive power.

The second study, with the same 91 participants described above, hypothesized that EI would mediate the association between family context (child maltreatment and parental mental health symptoms) and NSSI. Rates of maltreatment that met or exceeded the level indicating “low to moderate” abuse were relatively high in this clinical sample, with the highest rates reported for emotional forms of maltreatment (59.3% emotional abuse, 50.5% emotional neglect, 42.9% physical abuse, 34.1% physical neglect, 25.3% sexual abuse). Moreover, approximately 16.5% of parents reported clinically significant depressive symptoms and 22.0% reported clinical levels of global mental health impairment. Results did not support the hypothesized mediation model. Emotional neglect predicted the presence of NSSI one year later, but this effect was not mediated by EI. Further, no other forms of maltreatment (physical abuse or neglect, sexual abuse, emotional abuse) or parental mental health symptoms (depression, or global severity index) predicted NSSI over the course or a year, also suggesting a lack of mediation.

Across both studies, null results may be due in part to a relatively small sample size with a low reported rate of NSSI. Future research should continue exploring longitudinal ties between NSSI and coping-related factors such as EI, including in relation to family context. Understanding the directionality and specificity of the relation between NSSI, EI, and family factors could help clinicians target specific coping deficits and address family-related stressors that may influence this concerning behavior.