Interventions addressing mental health in conflict-affected settings: A literature review

The burden of mental health disorders is high in conflict-affected populations. However, there is still limited evidence from studies that measure the effectiveness of mental health interventions implemented in conflict and post-conflict settings. This study presents the results of a literature review aiming to identify those mental health interventions that have been implemented in such contexts. A systematic literature search was conducted in MEDLINE (PubMed), Scopus and PsycINFO leading to the identification of 33 relevant studies reporting on results of mental health-focused interventions with pre- and post- measurements. Eighteen studies were randomized clinical trials (RCTs), five were quasi-experimental and 10 were studies without controls. They covered 21 countries. Five RCTs reported effect sizes that were categorized as large. Their intervention models were Resource-Oriented Trauma Therapy (ROTATE), Problem Management Plus (PM+), a version of Narrative Exposure Therapy adapted for forensic offenders (FORNET), a contextual adaptation of the Unified Protocol (CXA-UP) and Common Elements Treatment Approach (CETA) respectively. Five other RCTs showed moderate effect sizes. All quasi-experimental studies and studies without controls reported some level of improvement in clients' well-being. Target groups and intervention types as well as providers of the interventions differed significantly across studies, as did outcome measures. PM + and NET including FORNET are the interventions best studied and have shown moderate to high effect in several contexts and for different populations. CETA and ROTATE have shown large effect in one RCT each. These four treatment modalities are all focused on the individual and delivered by mental health professionals, but some can also be delivered in groups and by lay health workers, which may be important in low resource contexts and where emphasis is put on the collective. Ways to standardize and manualize interventions need to be found so that they can be adequately tested in different contexts, on different well-defined target groups, with due consideration to who provides the services in question and with well-defined outcome measures both immediately post-intervention and at long-term follow-up.