Systematic Review and Meta-Analysis of Psychosocial Interventions For 6-12-Year-Old Children Who Have Been Forcibly Displaced

Objective(s): To identify: 1)psychosocial interventions conducted in middle childhood with forcibly displaced children; 2) the methodological quality of included studies; 3)intervention effectiveness and 4)characteristics of the included interventions.

Data Sources: CINAHL, Embase, Medline, PsycInfo, and supplementary literature. Limitations were age, English language, and human participants.

Study Selection: Inclusion criteria: 1)mean age of participants was 5 years -12 years 11 months; 2)forced displacement of the participants due to war/persecution; 3)the study included a psychosocial intervention; 4)psychosocial outcome measures used. No restrictions on study design. Two reviewers assessed the abstracts. Nineteen studies were included.

Data Extraction: Forms were created, piloted, and modified. One author completed extraction. A second author was consulted over uncertainties until consensus was reached. Study designs were classified using the National Health and Medical Research Council Hierarchy of Evidence. Methodological quality was determined using the QualSyst appraisal checklist. Intervention approaches were categorised using ‘Who is Where When doing What’ (4Ws) tool. Intervention effects were explored through meta-analysis.

Data Synthesis: Schools were the most common intervention setting, and teachers the most common intervention provider. Meta-analysis showed promising interventions were Narrative Exposure Therapy (moderate effect), and large effect sizes were found for Cognitive Behavioural Therapy, Child-Centred Play Therapy, Eye Movement Desensitization and Reprocessing, creative interventions. Unstructured play or education alone produced negligible effects. Combined within-group intervention effect for all studies was moderate (z(17)= 5.89, p < .000, Hedge's g = 0.75, 95%CI = 0.50–1.00). Prediction interval mean effect size was 0.75 (95% CI=0.50–1.00). The combined between-group intervention effect for all studies was negligible, and not statistically significant (Z(5)= 1.32, p = .19, Hedge's g = 0.15, 95%CI = −0.07–0.37). The prediction interval mean effect size was 0.15 (95% CI=0.07-0.37).

Conclusions: Psychological-based activities demonstrated effectiveness for symptom reduction. Future interventions should include strengths-based approaches. Further research is required using randomised control designs, greater sample sizes, and longitudinal data.

Author(s) Disclosures: The authors have no relevant financial or non-financial interests to disclose.