Identifying Mechanisms of Change in Dialectical Behavior Therapy for Substance Use Disorders (DBT-SUD): A Mixed Methods Examination of Mediators of Substance Use Reduction

Abstract

Substance use disorders (SUDs) often co-occur with other mental health disorders, such as borderline personality disorder (BPD), post-traumatic stress disorder (PTSD), and eating disorders. Historically, treatment systems have struggled to treat both substance use and other co-occurring mental health symptoms concurrently, often requiring abstinence before access to therapy is offered. Such approaches miss the mark in how to meet the needs of these complex individuals with co-occurring SUD and other mental health disorders. Dialectical Behavior Therapy for Substance Use Disorders (DBT-SUD) offers an integrated model tailored to address the needs of this population, combining evidence-based strategies to address emotion dysregulation and foster behavioral changes, such as moderation or abstinence from substances, as well as targeting suicidal and non-suicidal self-injurious behaviors. However, the mechanisms through which DBT-SUD achieve behavioral changes, particularly reductions in substance use, are not well understood. This study utilized an exploratory sequential mixed methods design to identify and test potential mechanisms of substance use change in DBT-SUD. Phase 1. Traditional ethnographic methods were used to analyze video-recorded therapy sessions from an RCT of DBT-SUD. Using both inductive and deductive coding strategies, four themes emerged that appeared to support reductions in substance use: reducing experiential avoidance, developing strong therapeutic alliance, cultivating self-validation and self-efficacy, and adopting dialectical thinking. Reduction in opioid use appeared to occur not through isolated skill use, but through dynamic interactions between therapist and client where trust and safety were paramount. This therapeutic alliance and safe environment seemed to allow for more willingness to experience emotions and engage in behavior change. Phase 2. Informed by the qualitative findings, this phase tested three longitudinal path models to examine whether these hypothesized mechanisms predicted reductions in opioid use across the treatment year. Data included repeated assessments of experiential avoidance, emotional ambivalence, emotion regulation, therapeutic alliance, and reduction in opioid use (via UA). While the hypothesized mediation pathways were not supported by the models, therapeutic alliance emerged as a significant predictor of reduction in opioid use and increased emotion regulation. These findings provide emerging evidence for the therapeutic alliance as a potential mechanism of change in DBT-SUD and underscore the need for more nuanced and precise ways of operationalizing alliance within complex treatments.