Preliminary Results of a Component Network Meta-Analysis for PTSD Psychotherapies

This study investigated the efficacy of various psychotherapeutic components for Post-Traumatic Stress Disorder (PTSD) using a component network meta-analysis (cNMA). A systematic literature search was conducted across multiple databases, including Cochrane, Embase, PsycINFO, PTSDpubs, PubMed, and CNKI, for peer-reviewed, full-text articles published between January 1st, 1980, and January 13th, 2025. Eligible randomized controlled trials (RCTs) included adult participants (at least 80%) diagnosed with clinical or subclinical PTSD. The primary outcome was post-treatment PTSD symptom severity, with secondary outcomes including follow-up PTSD severity and treatment drop-out rates. Standardised mean differences (SMDs) were used for continuous outcomes and odds ratios (ORs) for dichotomous outcomes. Both Bayesian cNMA and frequentist network meta-analysis were performed. An interaction term for bilateral stimulation and brief recalling was included in the primary model, and Bayesian LASSO was employed to manage other potential pairwise interactions. Out of 4258 records identified, 170 unique RCTs involving 13,157 participants were included. The cNMA results for post-treatment PTSD severity indicated that imagery rescripting and prolonged recalling of traumatic memory were beneficial. This analysis provides a nuanced understanding of the effectiveness of individual therapeutic components in PTSD treatment.