Structural interventions for HIV prevention are typically complex and evaluation should include measurement of implementation. In a trial of a structural intervention for HIV prevention in Botswana we trained teachers to use an audio-drama about gender violence and HIV (BVV). We measured the use of BVV by the trained teachers, and the factors related. In three-day workshops we trained guidance teachers in primary and secondary schools to use the 8-episode BVV audio-drama, that covers gender, gender violence and HIV. One to two years later, two interviewers visited schools and administered an electronic questionnaire to the head teacher and to the BVV-trained teacher. Most teachers (70%, 72/103) had used the BVV materials and reported a positive response from students. Primary school teachers were less likely to have used BVV (adjusted odds ratio (ORa) 0.24, 95% cluster adjusted confidence interval (CIca) 0.07-0.88). Teachers in schools with a working MP3 player were more likely to have used BVV (ORa 3.75, 95% CIca 1.11-12.70). Implementation was much lower in one district (ORa 0.12, 95% CIca 0.04-0.36), related to language constraints. The main difficulty was lack of time, especially in primary schools. These findings could inform rollout of the BVV programme in schools. Abbreviations: BVV: Beyond Victims and Villains audio-drama; CI: Confidence interval; OR: Odds ratio.
Keywords: HIV prevention; children; edutainment; implementation research; schools.