Abstract
IMPORTANCE: Childhood abuse significantly increases the risk of developing posttraumatic stress disorder (PTSD), often accompanied by symptoms of borderline personality disorder (BPD) and other co-occurring mental disorders. Despite the high prevalence, systematic evaluations of evidence-based treatments for PTSD after childhood abuse are sparse. OBJECTIVE: To compare the efficacy of dialectical behavior therapy for PTSD (DBT-PTSD), a new, specifically designed, phase-based treatment program, against that of cognitive processing therapy (CPT), one of the best empirically supported treatments for PTSD. DESIGN, SETTING, AND PARTICIPANTS: From January 2014 to October 2016, women who sought treatment were included in a multicenter randomized clinical trial with blinded outcome assessments at 3 German university outpatient clinics. The participants were prospectively observed for 15 months. Women with childhood abuse-associated PTSD who additionally met 3 or more DSM-5 criteria for BPD, including affective instability, were included. Data analysis took place from October 2018 to December 2019. INTERVENTIONS: Participants received equal dosages and frequencies of DBT-PTSD or CPT, up to 45 individual sessions within 1 year and 3 additional sessions during the following 3 months. MAIN OUTCOMES AND MEASURES: The predefined primary outcome was the course of the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5) score from randomization to month 15. Intent-to-treat analyses based on dimensional CAPS-5 scores were complemented by categorical outcome measures assessing symptomatic remission, reliable improvement, and reliable recovery. RESULTS: Of 955 consecutive individuals assessed for eligibility, 193 were randomized (DBT-PTSD, 98; CPT, 95; mean [SD] age, 36.3 [11.1] years) and included in the intent-to-treat analyses. Analysis revealed significantly improved CAPS-5 scores in both groups (effect sizes: DBT-PTSD: d, 1.35; CPT: d, 0.98) and a small but statistically significant superiority of DBT-PTSD (group difference: 4.82 [95% CI, 0.67-8.96]; P = .02; d, 0.33). Compared with the CPT group, participants in the DBT-PTSD group were less likely to drop out early (37 [39.0%] vs 25 [25.5%]; P = .046) and had higher rates of symptomatic remission (35 [40.7%] vs 52 [58.4%]; P = .02), reliable improvement (53 [55.8%] vs 73 [74.5%]; P = .006), and reliable recovery (34 [38.6%] vs 52 [57.1%]; P = .01). CONCLUSIONS AND RELEVANCE: These findings support the efficacy of DBT-PTSD and CPT in the treatment of women with childhood abuse-associated complex PTSD. Results pertaining to the primary outcomes favored DBT-PTSD. The study shows that even severe childhood abuse-associated PTSD with emotion dysregulation can be treated efficaciously. TRIAL REGISTRATION: German Clinical Trials Register: DRKS00005578.