Intimate partner violence is highly

Intimate partner violence is highly prevalent in sub-Saharan Africa and its role in increasing women\'s risk of HIV acquisition is well recognised, although many aspects of the mechanisms through which HIV increases women\'s risk remain unclear. The most important challenge for researchers is to develop combination interventions that effectively reduce both the risk of HIV acquisition and the risk of intimate partner violence. Thus far there have been key trials in South Africa with the Stepping Stones intervention and the IMAGES study, which combine microfinance and a gender intervention. However, both interventions reduced intimate partner violence but had no effect on HIV acquisition. In , Jennifer Wagman and colleagues report findings of their SHARE study in Rakai, Uganda—the first behavioural intervention to reduce both HIV incidence and intimate partner violence reported in women. SHARE is a multi-component intervention that involves mainly CHC action to change of social norms that support violence and to provide enhanced HIV counselling, including information on intimate partner violence and safety for women who test positive for HIV. It has been hypothesised that community-wide interventions that focus on social norm change are an important next direction for prevention of violence against women because, by definition, they intervene upstream and are positioned to change the conditions that might undermine sustainability of impact. An assessment of Sasa!, a similar intervention in Uganda, did not show significant effect on physical intimate partner violence, but this finding was probably because of insufficient power. Wagman and colleagues\' findings provide important support for an intervention model. Although social norm change was not the only component of SHARE and the effect of enhanced counselling on women\'s risk of intimate partner violence is not known, it is hard to imagine that it was an important driver of the main finding given that only 56 women in the intervention communities in the study had seroconverted and so might have been using services. The intervention did not have a significant effect on self-reported male perpetration of intimate partner violence. Men could have been less honest in their disclosure of perpetration than were women, and this difference might have made their reports unreliable as an outcome. However, previous population-based studies generally note that women\'s reports of intimate partner violence tend to be the same as men or lower in prevalence than men\'s of perpetration, but perhaps what is seen here is a trial effect. Assessments of interventions against violence have an inherent weakness in that biological measures do not exist to valuate self-reports. In view of the stressfull effect of home exposure to violence, a future development in the assessment of interventions against intimate partner violence should include biological markers of stress, (eg, cortisol) as an indicator of effect. Compared with individuals in control clusters given only standard HIV services, individuals in the SHARE intervention groups reported significantly less past-year physical (346 [16%] 217 [12%] in intervention groups; adjusted prevalence risk ratio 0·79, 95% CI 0·67–0·92) and sexual (261 [13%] 167 [10%]; 0·80, 0·67–0·97) intimate partner violence after 24 months. Findings were not significant at 12 months (follow-up 1). These findings support those of Stepping Stones Study and IMAGE trial, both of which show that the effects of major interventions evolve with interventions rather than attenuate. This difference shows the need for researchers and donors to be mindful of the need for longer term follow-up in these trials to avoid prematurely negative findings. The evidence base for prevention interventions for intimate partner violence and dual programming to reduce HIV acquisition in sub-Saharan Africa is still very small. Assessments need to be replicated and translation research carried out into the conditions and inputs required to scale-up effects, and its cost and effectiveness. To advance, the specialty needs injections of donor funds to replicate proven interventions and increase evidence. In this regard, the Department for International Development global programme could escalate the science of prevention of intimate partner violence and evidence of intervention outcomes in several regions worldwide. However, substantially more large-scale and long-term investment is needed to advance towards a goal of ending violence against women and girls.