Characteristics of participants in an HIV prevention intervention for youth in Rwanda: results from a longitudinal study

Abstract

Introduction. This paper studies determinants of participation in a peer-led school-based HIV prevention intervention in Rwanda. Methods. A baseline survey among 1071 students (mean age 17 years) assessed potential determinants of participation, while a follow-up six months in the intervention measured actual participation in the intervention. Statistical models were built using multivariate linear and multinomial regression analysis predicting overall participation, par- ticipation in group discussions and individual counseling. Results. Those who recently had sex, had been tested for HIV, feel more susceptible to HIV, have a higher sexual self-concept, a more positive future perspective (only for non-sexually active), and boys, were more likely to participate in group activities. Also students from the same class as the peer educator and boarding school students were more likely to participate in group activities. Older students and those with low external health locus of control participated more in individual counseling. Discussion. Participation could be increased by investing in general well-being of young people, organizing girls-only activities, and diversifying activities. Key words: selection bias, HIV prevention, participation rate, young people, Rwanda