Preventing violence against conflict-affected adolescent girls: findings from Ethiopia, Democratic Republic of Congo, and Pakistan

Abstract Background Interpersonal violence is a major public health concern in humanitarian contexts, and adolescent girls are uniquely vulnerable. There is little evidence on effective violence prevention programmes targeting this population. We investigated the efficacy, feasibility, and acceptability of a life skills and safe spaces programme to reduce violence exposure for conflict-affected adolescent girls. Methods Evaluations were conducted in Ethiopia, the Democratic Republic of Congo (DRC), and Pakistan. In Ethiopia and DRC, the programme was evaluated via cluster-randomised controlled trials in refugee camps. In Ethiopia, girls aged 13–19 were enrolled. Girls and their caregivers in the treatment group received the intervention and the control group was waitlisted. In DRC, girls aged 10–14 and their caregivers were enrolled. All girls received the intervention and treatment was randomised for caregivers. In both Ethiopia and DRC, the primary outcome was girls' exposure to sexual violence. In Pakistan, a mixed-methods evaluation design focused on feasibility and acceptability. A single-group within-participants pre-post test was administered to 78 internally displaced and host community girls aged 12–19 enrolled in the programme. Qualitative in-depth interviews were conducted with a subset of 15 girls at endline. Primary outcomes included perceptions of programme content and changes in freedom of movement, safety, self-esteem, and hope. Findings In Ethiopia, 919 girls were enrolled, 457 in the intervention group and 462 in the wait-list group. At 12-month follow-up, the intervention was not significantly associated with reduction in exposure to sexual violence (aOR=0·96, 95% CI 0·59–1·57). In DRC, 869 girls ages 10–14 and 764 caregivers were enrolled. 389 caregivers received the intervention and 375 were wait-listed. At 12-month follow-up, the intervention was not significantly associated with reduction in exposure to sexual violence (aOR=0·95, 95% CI 0·65–1·37). In Pakistan, programme content was perceived as acceptable and freedom of movement, safety, self-esteem, and hope increased over time. Interpretation The intervention was feasible and acceptable in Pakistan, but further research and programmatic adaptations are needed to prevent adolescent violence in humanitarian contexts. Funding UK Department for International Development.