Measuring and Modifying Potential Mechanisms of Change in Self-Injury

Non-suicidal self-injury (NSSI) is a prevalent and dangerous behavior. Research has shown that people who engage in NSSI have deficits in emotion regulation and problem solving skills and that this behavior is most often performed as means of compensating for these deficits. Most treatments for NSSI have focused on improving these two skills, but none has shown efficacy, raising the question of whether these are malleable skills among self-injurers. This study tested the malleability of these two skills using a brief laboratory manipulation. We also tested whether change in these skills statistically mediated behavior change at two-week follow-up. Finally, we tested potential moderators of these effects.;Participants were 48 adults recruited from the community who actively engaged in NSSI. All participants completed a diagnostic interview and behavioral assessment; two behavioral tasks measuring emotion regulation and social problem-solving skills, respectively; one of three approximately one hour manipulations focused on improving emotion regulation, problem-solving skills, or a psychoeducation control condition; and then completed the two behavioral tasks again. All participants were then contacted by telephone two weeks later and their self-injurious behavior assessed.;Results indicated that many aspects of these skills improved regardless of manipulation condition. Those who received the emotion regulation manipulation showed decreased performance on the task designed to measure emotion regulation skills and those who received the problem-solving manipulation showed improvement on the task designed to measure social problem-solving skills. The manipulations did not affect self-injurious thoughts and behaviors. Furthermore, change in the skills measured did not significantly mediate the relation between the manipulation received and change in self-injurious behavior. Finally, participants' reported reasons for engaging in NSSI significantly moderated the effects of the manipulations; however, diagnosis of borderline personality disorder (BPD) did not.;This study provides initial support for the feasibility of using behavioral tasks to measure mechanisms through which NSSI may be maintained and the possibility that individual-level variables might impact the effects of treatment. Future research is needed to further investigate the mechanisms and moderators of change in treatment for NSSI to create the most effective and efficient interventions.