Lack of evidence for the efficacy of psychotherapies for PTSD and depression in child and adolescent refugees

Post‐traumatic stress disorder (PTSD) and depression are known to be prevalent among young refugees and internally displaced individuals. The need for effective interventions for this population is becoming increasingly evident in view of the large number of recent and current armed conflicts.

We conducted a systematic review and meta‐analysis of randomized controlled trials on the efficacy of psychological interventions for PTSD and depression in child and adolescent refugees and internally displaced individuals. The aims and methods of the meta‐analysis were registered with the PROSPERO database (CRD42017071384).

We searched the databases PsycINFO, MEDLINE, and PILOTS up to February 2018. The following search terms in keywords, titles and abstracts were used: PTSD, depression, refugee (or related terms such as asylum seeker or displaced person), and treatment (or related terms such as intervention or psychotherapy). Additionally, reference lists of identified publications and systematic reviews were examined.

The inclusion criteria were: a) trial conducted with child or adolescent refugees or internally displaced individuals; b) participants randomly assigned to treatment conditions; c) at least ten participants completing an active psychological treatment for PTSD or depression or both. No restrictions were made upon intervention format, publication type, or publication language. If studies did not provide sufficient data for performing the meta‐analysis, the authors were contacted by e‐mail to retrieve these data.

We coded and extracted relevant study, intervention and participant characteristics, such as number of participants, comparison group(s), type of outcome measure used, outcome scores, and number of sessions. Furthermore, we rated the quality of the included trials by applying nine criteria used in similar meta‐analyses1. To conduct the analyses, the control group mean was subtracted from the treatment group mean at post‐treatment or follow‐up, and divided by the pooled standard deviation. Subsequently, to obtain the effect size Hedges's g, the outcome was multiplied by a sample size correction factor and the random effects model was applied. Analyses were completed with Comprehensive Meta‐Analysis Version 3. Given that less than ten trials met our inclusion criteria, no test of publication bias could be conducted.

After screening 1,716 potential hits, eight trials met our criteria2, 3, 4, 5, 6, 7, 8, 9. All publications were written in English, seven were published in peer‐reviewed journals and one was a doctoral thesis7. Seven of the trials were conducted with internally displaced individuals, whereas two were conducted with refugees6, 8. In three trials, treatment was performed in group format2, 5, 9. Four trials assessed both PTSD and depression4, 5, 7, 9, one focused on depression only2, and three focused on PTSD only3, 6, 8.

Experimental conditions consisted of trauma‐focused cognitive behavior therapy that included narrative exposure therapy3, 4, 6, 7, 8, interpersonal therapy2, classroom‐based intervention9, and writing for recovery5. Active treatments were compared to waitlist in four trials. In two trials, the experimental condition was compared to an inactive control condition in addition to the waitlist. In three trials, two active conditions were compared to each other.

The number of participants per condition varied from 11 to 248, with a mean of 78.7±61.9. The mean age of participants was 13.1±1.9, and 49.9% of them were female. Two and three trails, respectively, used structured clinical interviews to assess PTSD or depression; the remaining trials applied self‐reports. The number of sessions ranged from 6 to 16.

Active treatments for PTSD yielded non‐significant aggregated effect sizes at post‐treatment (k=7; g=0.02; 95% CI: −0.13 to 0.15) and at follow‐up (k=5; g=0.24; 95% CI: −0.13 to 0.62) when compared to control conditions. Only one trial produced a significant effect of the experimental condition over the control condition6.

Active treatments for depression also produced non‐significant aggregated effect sizes at post‐treatment (k=6; g=−0.01; 95% CI: −0.55 to 0.52) and at follow‐up (k=3; g=0.02; 95% CI: −0.16 to 0.19) when compared to control conditions. Only one trial showed a large effect for depression3.

Three trials that reported on functional impairment led to a small effect size at post‐treatment (k=3; g=0.31; 95% CI: 0.08‐0.54) and at follow‐up (k=3; g=0.32; 95% CI: 0.01‐0.64) when active treatments were compared to control conditions.

The assessment of quality of the included publications indicated that six trials (75%) were rated with an average score of 2, indicating good quality.

This meta‐analysis demonstrates that there is a limited number of clinical trials on the efficacy of psychotherapies for PTSD and depression among child and adolescent refugees and internally displaced individuals. The results of existing trials do not provide support for the efficacy of psychological interventions in this population.

Given the urgent public health issues raised by escalating levels of violence and civil conflict around the globe, it is essential for government and non‐government agencies to have the most reliable evidence to shape policy and practice. Accordingly, we urgently need to develop and test effective interventions for mental health problems in young refugees and internally displaced individuals.