Adherence and competence in two manual-guided therapies for co-occurring substance use and posttraumatic stress disorders: clinician factors and patient outcomes

Background: The challenges of implementing and sustaining evidence-based therapies into routine practice have been well-documented. Objectives: This study examines the relationship among clinician factors, quality of therapy delivery, and patient outcomes. Methods: Within a randomized controlled trial, 121 patients with current co-occurring substance use and posttraumatic stress disorders were allocated to receive either manualized Integrated Cognitive Behavioral Therapy (ICBT) or Individual Addiction Counseling (IAC). Twenty-two clinicians from seven addiction treatment programs were trained and supervised to deliver both therapies. Clinician characteristics were assessed at baseline; clinician adherence and competence were assessed over the course of delivering both therapies; and patient outcomes were measured at baseline and 6-month follow-up. Results: Although ICBT was delivered at acceptable levels, clinicians were significantly more adherent to IAC (p<0.05). At session 1, clinical female gender (p<0.05) and lower education level (p<0.05) were predictive of increased clinician adherence and competence across both therapies. Adherence and competence at session 1 in either therapy were significantly predictive of positive patient outcomes. ICBT adherence (p<0.05) and competence (p<0.01) were predictive of PTSD symptom reduction, whereas IAC adherence (p<0.01) and competence (p<0.01) were associated with decreased drug problem severity. Conclusions: The differential impact of adherence and competence for both therapy types is consistent with their purported primary target: ICBT for PTSD and IAC for substance use. These findings also suggest the benefits of considering clinician factors when implementing manual-guided therapies. Future research should focus on diverse clinician samples, randomization of clinicians to therapy type, and prospective designs to evaluate models of supervision and quality monitoring.