Despite the importance of preventing HIV infection in young people in countries where it is highly endemic, few rigorously conducted studies have measured the effects of preventive interventions.1 The linked randomised trial by Jewkes and colleagues (doi: 10.1136/bmj.a506) of the Stepping Stones intervention is therefore an important addition.2 The trial randomly allocated 70 villages in the Eastern Cape province of South Africa to the intervention arm or control arm. In each village, 20 male and 20 female volunteers took part in Stepping Stones, a 50 hour programme to improve sexual health using participatory learning approaches, or a three hour control programme. After two years of follow-up, the incidence of HIV (the primary outcome) was not significantly affected, but infection with herpes simplex type 2 virus (HSV-2) was significantly (33%) reduced. In young men, reported risk behaviours, perpetration of intimate partner violence, and problem drinking were also significantly reduced, but no differences were seen in young …