Predicting the Success of Skill Training on Managing Emotions

Abstract

Borderline Personality Disorder (BPD) is a highly distressing disorder that affects millions of people in the United States. Notoriously difficult to treat, BPD is characterized by impairments in self and interpersonal functioning, negative affectivity, disinhibition, antagonism, and high rates of suicide. Dialectic Behavior Therapy (DBT)–one of the main therapies developed to treat BPD–views emotion dysregulation as the main cause of borderline symptoms. To this day, treatment fails a significant percentage of borderline patients. We suggest improving treatment through analysis of individual differences. The current study examined moderators of DBT skill training. Participants who scored above average on the Difficulties in Emotion Regulation Scale (DERS) (n = 133) were randomly assigned to two groups, with Radical Acceptance–a distress tolerance skill from DBT–taught to the Skill Learning group, and general information surrounding emotion regulation taught to the Education Control group. Treatment outcome was measured in terms of self-reported anger and assigned punishment, after participants played a series of games meant to frustrate them. Both measures revealed how well participants regulated their emotions. A series of regressions were conducted to test for main effects and interaction concerning five hypotheses. On average, Education Control participants reported more anger than Skill Learning participants, but no significant differences between groups were found in average punishment, and in positive and negative affect. Agreeableness and two DERS subscales, Nonacceptance and Clarity, were found to moderate intervention outcome. The scarcity of significant results could in part be attributed to the lack of power. Future studies should include all DERS scores so as to not limit the range of the predictor variable, and increase the sample size to capture the small difference in the dependent measures hidden by the floor effect. Moreover, pre- and post-measurements for the dependent variables are recommended for future studies, as they could further reveal the mechanism behind any difference between groups.