Narrative exposure therapy for posttraumatic stress disorder: A meta-analysis of randomized controlled trials

Objective: Narrative exposure therapy (NET) is a psychological intervention conditionally recommended for the prevention and intervention of posttraumatic stress disorder (PTSD). Previous study reported that NET had medium effect on PTSD, but the evidence base of NET is still weak. In this article, we conducted a meta-analysis to explore the efficacy of NET for PTSD symptom reduction and loss of diagnosis. Method: Databases were searched for randomized controlled trials of NET for PTSD. Effect sizes were calculated by Hedges’ g and ratio of risk (RR) with 95% confidence intervals (CIs). Study heterogeneity was assessed by Q, Tau2 and I2 and explored by subgroup analyses and metaregression analyses. Results: 18 studies which met full inclusion criteria were included. The finding showed that NET had moderate between-group effect size (g = .57, 95% CI [0.87, 0.28], p < .01) and large within-group effect size (g = 1.31, [1.54, 1.09], p < .01) at first postintervention assessment. The mean RR of loss of diagnosis between NET and comparators was 2.20 ([1.23, 3.92], p < .01). However, there were no differences between groups in both symptom reduction and loss of diagnosis when accounting for publication bias. And the heterogeneity could not be explained sufficiently. Conclusion: Although the article indicated the efficacy of NET for PTSD symptom reduction and the loss of diagnosis, it is difficult to draw meaningful conclusions considering the heterogeneity and publication bias in samples. Impact Statement: The global burden of posttraumatic stress disorder (PTSD) is a challenging problem because of the growing refugee population. Narrative exposure treatment (NET) is a kind of treatment for PTSD, especially among refugees. However, it still needs more evidence to assess the efficacy of NET for PTSD. This study was conducted to explore the efficacy for PTSD symptom reduction and loss of diagnosis. Results indicated medium effect sizes of NET but there were no differences between groups when accounting for publication bias. The findings are informative for mental health professionals and related organizations to consider the application of NET.