The role of gender-based violence in fuelling the global HIV pandemic is now undeniable. In recent years, research from both developed and developing countries has consistently shown that women who experience gender-based violence and gender inequality are at greater risk of HIV.^1–4 Mounting evidence on connections between men’s perpetration of gender-based violence and male HIV risk behaviour suggests that women’s HIV risk is often primarily dependent on the behaviour of the men in their lives. Research from South Africa, India and the United States has suggested that men who are violent towards their female partners or commit rape are more likely to have sex more often, to have sex with concurrent and/or casual sexual partners, to have higher total numbers of sexual partners, to practice anal sex, to participate in transactional sex, to father children, and to use alcohol and drugs.^5–9

In this issue of Sexually Transmitted Infections, Jay Silverman and colleagues¹⁰ present findings from Bangladesh which show that men who have perpetrated physical or sexual violence against their wives in the past 12 months are more likely to report extramarital sex and STI symptoms or diagnosis within the same period; they are also more likely to report prior pre-marital sex. This paper offers the first evidence from a nationally representative sample of men that supports links between perpetration of gender-based violence and STI/HIV risk, and it expands again the range of cultural settings in which such associations have been observed. The national generalisability of Silverman et al’s findings, combined with the notable consistency of associations across widely diverse study populations,^5–9 affirms the fact that abused women’s sexual health is often placed at risk by their partner’s sexual behaviour. In light of such emerging knowledge, we suggest that continued advocacy of traditional individual-level ABC (Abstain, Be faithful, use Condoms) prevention messages for women is not only futile but morally bankrupt. Abstinence and condom use may be impossible for women to enforce. Fidelity is of no use unless it is mutual, and men’s faithfulness very often lies outside of women’s control.¹

To intervene effectively, however, we must understand the reasons for observed associations between violence perpetration and sexual risk taking among men. We suggest that both violence perpetration and sexual risk taking arise from a common underlying cause, and that this cause is social ideals of masculinity. Qualitative evidence from South Africa and other settings suggests that “successful” performance of masculinity often depends on both heterosexual success with women and, in the context of entrenched gender hierarchy, on the ability to control women.^11–13 There is also evidence that men’s controlling behavior can place women at increased risk of HIV infection, even in the absence of overt violence. For example, research in Soweto, South Africa showed that women who had less power in their current sexual relationship were more likely to have HIV, regardless of co-occurrence of physical or sexual abuse.² We suggest that effective interventions to protect both men’s and women’s sexual health will thus require not only interruption of intertwined cycles of violence perpetration and sexual risk taking among men but active transformation of underlying gender norms that legitimate male power, male control, male violence and men’s sexual risk taking.

This is why even gender-sensitive technology-based prevention approaches such as microbicides, while valuable, will ultimately address only part of the problem; they may address the immediate constrained realities of women’s lives but do not address the underlying social constructions of gender that create the constraints in the first place. Broader transformative programs that link promoting gender equality and preventing violence perpetration and sexual risk taking are needed. Work with men needs to challenge ideas about gender relations that construct women as objects of sexual conquest, and thereby encourage multiple partnering and legitimise the use of violence in controlling female partners. We must also enhance the agency that women do have in sexual decision making and risk reduction, through reflecting critically on the origins of HIV risk in gender and often socio-economic inequality, challenging social sources of inequality and empowering women to make different choices. To the extent that many transmissions of HIV are believed to occur within established relationships,¹⁴ it is similarly crucial to promote gender equality and better communication in heterosexual relationships.

The global HIV pandemic in its current form cannot be effectively arrested without fundamental transformation of gender norms. The scale of change needed may seem overwhelming, and the need for cultural sensitivity in creating prevention strategies may seem to some to be fundamentally at odds with an explicit agenda for cultural transformation. Nonetheless, there are promising strategies on the horizon. Stepping Stones is an often-cited example of a gender-transformative program. Developed in Uganda, it has since been used in over 40 countries and translated into at least 13 languages.¹⁵ Forthcoming findings from a community-based randomised controlled trial of Stepping Stones in South Africa show that over 2 years of follow-up, there was a sustained reduction in young men’s perpetration of intimate partner violence, as well as in their number of partners and participation in transactional sex, and an increase in condom use.¹⁶ In the absence of a vaccine, which is not anticipated for many years to come, the real prize in terms of sustainable, long-term HIV prevention may lie in successfully creating and disseminating programs which promote such changes.