Rethinking power: secondary analysis of evaluation results for an intervention to prevent violence against adolescent girls and young women in Haiti.

Background: Though there are a growing number of programs seeking to prevent violence against women and children, adolescent girls often fall into a gap between these approaches. This article focuses on the impact of a violence prevention program, Rethinking Power, on the lives and wellbeing girls and young women aged 10-23.

Methods: The study utilized a quasi-experimental, mixed-methods design. The results focus on data collected with girls and young women (N = 1,627) who were of adolescent age during program implementation. A difference-in-difference approach explored the impact of the program on key outcomes using cross-sectional data from three timepoints. A survey of girls' groups participants was also undertaken (N = 752) and qualitative data was collected with adolescents and community stakeholders (52 focus groups and 61 interviews). Regression was utilized to analyze girls' group data and thematic analysis to analyze qualitative data. Data for mixed methods analysis was brought together using a convergent approach.

Results: The results show girls and young women the intervention areas reported less experiences of physical or sexual IPV in the past 12 months over time (from 25.7% at baseline to 15.7% at endline) and reduced, though non-significant due to small samples, odds of experiencing IPV compared to the controls (OR: 0.78; p = .51). Qualitative data also showed changes in violence, acceptance of violence and gender attitudes. Girls and young women in the intervention area (rather than control) had more than twice the odds of reporting that they could choose who to be friends with (OR: 2.59; p = .047) and had greater odds of agreeing that girls should be allowed to socialize just as boys (OR: 3.87; p <.001) and that a man should not have the final word at home (OR: 1.94; p = .019). Participants in girls' groups saw improvements on indicators related to gender attitudes, acceptance of violence, self-esteem and agency.

Conclusions: While not all results are statistically significant due to small sample sizes, they suggest that girl-focused programming, alongside a wider community-change process, can improve the lives of adolescent girls.

Trial Registration: The full trial was retrospectively registered in ISRCTN (ISRCTN12311597) on April 4th, 2024.