Intended for healthcare professionals


Partner reduction and the prevention of HIV/AIDS

BMJ 2004; 328 doi: (Published 08 April 2004) Cite this as: BMJ 2004;328:848
  1. David Wilson, senior monitoring and evaluation specialist (dwilson{at}
  1. Global HIV/AIDS Program, World Bank, 1818 H Street N.W., Washington, DC 20433, USA

    The most effective strategies come from within communities

    In an era of increasingly complex HIV/AIDS analyses and responses, Shelton et al reaffirm the simple truth that without multiple sexual partnerships, an HIV epidemic would not occur and that by extension partner reduction is the most obvious, yet paradoxically neglected, approach to the prevention of HIV (p 891).1 They note that in the ABC model for preventing AIDS/HIV (abstinence, or deferred sexual inception—A, be faithful, or partner reduction—B, and condom use—C), sexual deferral and condom use have persuasive advocates but partner reduction does not.

    Their analysis of the vital part played by partner reduction in reducing HIV infection in Western gay communities, Uganda, and Thailand is timely. We face a crisis in HIV prevention. The successes in Uganda and Thailand occurred 15 years ago, and in the intervening period no national declines of similar clarity or scope have occurred. Similarly, in HIV prevention research, the heady days of the Mwanza sexually transmitted infections trial were succeeded by the disappointing findings (albeit explicable) in the more ambitious Rakai sexually transmitted infections trial, the Masaka triplet IEC (information, education, and communication) and sexually transmitted infections trial, and most distressingly, the recent Mwanza adolescent trial.25 Shelton et al's analysis may help to infuse new life into HIV/AIDS prevention. Their argument that partner reduction is the potential centre-piece of a unified ABC approach is good common sense—and good epidemiology.

    Whether the ABC approach addresses the needs of women is debatable, with commentators arguing that many women are unable to negotiate relationships based on abstinence, faithfulness, or condom use.6 The enduring contribution of gender inequalities, including economic inequality and gender violence, to women's vulnerability to HIV is incontrovertible. Yet it is intriguing that some of the steepest declines in HIV infection levels in Uganda seem to have occurred among women, particularly young women, putatively the most powerless members of society. Shelton et al present evidence that where HIV prevalence has declined among pregnant women (Uganda, Thailand, Zambia, Ethiopia, Cambodia, and the Dominican Republic) the primary reported behaviour change has been partner reduction and monogamy by men, especially older men. Uganda's experience shows that achieving sexual deferral and partner reduction among men, particularly older men, may create safer environments for women, particularly young women. Community norms that proscribe older men having sexual relationships with younger women may be especially protective. A successful ABC approach that reduces HIV infection among women, particularly young women, is a vital element of a broader gender response. Uganda's ABC approach was reinforced by practical measures to increase women's participation in higher education and political life and to protect women from gender violence and sexual coercion.

    Analysis of factors contributing to behaviour change in Uganda and elsewhere is even more challenging than the reaffirmation of partner reduction. Contexts as disparate as California, Uganda, and Thailand share unnerving similarities.710 Above all, HIV prevention responses were rapid, endogenous, inexpensive, and simple.8 9 They were based on the premise that communities, however disparate, have within themselves the resources and capital to reverse this epidemic. They preceded large scale exogenous assistance and occurred largely without the involvement of specialist agencies. They were locally led, by gay leaders and activists in California and by political, religious, and community leaders in Uganda. They promoted changes in community norms, thus creating enabling and protective environments long before the concept gained currency. They stressed simple messages and actions and in doing so achieved declines in HIV infection that preceded the growth in HIV services, including distribution of condoms and voluntary counselling and testing. They relied on interpersonal communication channels and networks, rather than mass media.8 9 11

    Remarkably they combined high fear approaches with openness and the capacity to rise above discrimination and to integrate prevention and care effectively.8 9 In doing so they created a context in which people perceived high personal risk of HIV infection and a personal proximity to the epidemic (measured, for example, by the extent to which we know people who have died of AIDS) that many communities with equally high HIV infection levels have not yet attained. Despite our lament that behaviour change is slow, they achieved rapid declines in risky sexual behaviour and HIV infection. The slowest element was our capacity to recognise the rapidity and extent of these changes. They unified personal values and societal messages to achieve conviction and consistency. As AIDS educators, we often publicly promote approaches that we would not countenance in our own personal lives, such as the notion that it is acceptable for our spouses or children to have multiple partners, provided condoms are used. In Uganda, emphasis on the primacy of partner reduction resonated with community perspectives.

    Partner reduction is good epidemiology, not good ideology, and we must ensure that the ABC approach remains sufficiently scientifically grounded to withstand shifting ideological sands. Happily epidemiology's insights are diverse enough to affront all our ideologies in equal measure. While Uganda's achievements imply a major role for partner reduction, data from, for example, Nairobi, Abidjan, Accra, many other cities in Africa, and large swathes of Asia support a major focus on making sex work safe, through rights based legal reform, enhanced sexual health care, and promotion of condoms. Similarly, the epidemics in the former Soviet Union and much of Asia cry out for a major commitment to comprehensive initiatives to reduce harm to injecting drug users.

    We are indebted to Shelton et al for calling attention to the importance of partner reduction and its possible determinants and the implications for our programmes. We must also recognise that many communities have not developed similarly effective local responses, and respond with improved epidemiological and social research to ensure we understand what happened in Uganda and elsewhere. We require this to communicate persuasively with hesitant communities and to improve our ability to facilitate and nurture effective local responses. In short, we must foster endogenous responses founded primarily on the resources, capital, and leadership within communities while enhancing research to ensure these responses are understood, evaluated, and illuminated by science.


    • Education and debate p 891

    • The findings, interpretations, and conclusions expressed in this paper are entirely those of the author. They do not necessarily represent the view of the World Bank, its executive directors, or the countries they represent

    • Competing interests None declared.


    View Abstract