Feature HIV and AIDS

AIDS and the irrational

BMJ 2008; 337 doi: http://dx.doi.org/10.1136/bmj.a2638 (Published 26 November 2008) Cite this as: BMJ 2008;337:a2638
  1. Helen Epstein, independent consultant on public health in developing countries
  1. 1New York NY 10031, USA
  1. helenepstein{at}yahoo.com

    The director of UNAIDS, Peter Piot, is stepping down in the next few weeks. In an open letter to his replacement, Helen Epstein argues that the agency should give the issue of concurrent partners high priority in combating HIV in Africa

    In a recent survey of HIV positive South Africans, almost half believed that traditional African medicine is more effective than antiretroviral drugs.1 This is upsetting news. The country has invested heavily in antiretroviral drugs, rapid HIV tests, CD4 cell counters, and condoms and is the site of many clinical trials into novel treatments and HIV prevention devices. In the midst of all this technology, why do irrational beliefs about AIDS persist?

    The reasons are complex. AIDS advocacy groups attribute misconceptions about AIDS to sexual shame and the misguided leadership of former president Thabo Mbeki, who questioned the relation between HIV and AIDS. However, rumours about AIDS—that it is caused by witchcraft, US backed germ warfare against black people, or some foodborne poison—are common everywhere.2 In Nigeria, for example, a barber recently told a reporter that three quarters of his clients bring their own clippers because of fear of AIDS—even though there has never been a documented case of HIV transmission through hair dressing.3

    Unfortunately, the international public health community, and the Joint United Nations Programme on HIV/AIDS (UNAIDS) in particular, may be contributing to the mystification of AIDS in Africa by promoting a needlessly overcomplicated view of the epidemic that has sown confusion among researchers and ordinary Africans alike. Although UNAIDS can be proud of its success in getting most governments around the world to face up to AIDS, it must also face up to its failure to help the most severely affected communities understand the causes of the epidemic.

    The root of the problem may lie in the agency’s mandate. UNAIDS is both the most trusted source of scientific information on the global epidemic and an adviser to governments about how to tackle it. Unfortunately, these two goals are seldom compatible, especially given the heated competition over the vast amounts of money now devoted to AIDS.4

    The intrinsic tension between politics and science has been especially acute when it comes to answering two of the most vital questions in AIDS prevention: Why is the epidemic in Africa so severe? And what are the best ways of dealing with it? Although it is difficult to generalise, UNAIDS’s official explanation for the African AIDS epidemic is that it is due to many complex factors, including structural and gender inequalities, migration patterns, urbanisation, and other influences on the size and sexual behaviour of so called core groups—that is, sex workers and their typical clients (truck drivers, mine workers, soldiers, traders, and other migrant men).5 6 However, studies show that age adjusted HIV infection rates in southern Africa are nearly as high in the general population as they are among sex workers and migrant labourers.7 8 9 Furthermore, surveys from across the continent find that lifetime numbers of sexual partners in African countries tend to be similar to those in many Western countries,10 and much lower than in many countries in Asia, where formal prostitution is far more common.11 12 13 Yet HIV prevalence is orders of magnitude greater in southern Africa than in either Asia or the West.

    This paradox is apparent not only to AIDS researchers, but also to people who have watched numerous non-promiscuous friends and relatives succumb to the disease. This may partly underlie the general mistrust of official AIDS information and a search for other explanations. It may also help explain why so many people in Africa still do so poorly in surveys of AIDS related knowledge—especially when it comes to answering questions about HIV transmission through casual contact such as barber’s shears. The problem may not be lack of knowledge but disbelief about what they have been told about so called “risky” behaviour.

    Patterns of sexual behaviour

    Multiple concurrent partnerships provide a compelling, if partial, resolution of the apparent paradox of Africa’s high HIV infection rates.14 Although African people may not have more sexual partners than those in other countries, they are more likely to have two or three long term concurrent partners (reference box). This pattern of behaviour gives rise to an interlocking network of sexual relationships that creates a superhighway for HIV. If one person is infected, everyone is at high risk, including those with only a few long term partners, or even only one (box ).15 Casual and commercial sex remain important risk factors, but “long term concurrency” probably explains why HIV in Africa has spread so rapidly beyond typical “high risk groups” such as sex workers.

    Risks of concurrency compared with serial monogamy

    If a man has two partners, his risk is the same whether they are concurrent or serial. However, his concurrency hugely affects his partners’ risks, because both partners are now linked through him, and thus at risk not only from him directly, but from each other, indirectly too (figure). If one of his partners is infected, the other will become infected right away, if they are concurrent. If he has those partners sequentially, the infection spreads much more slowly from one woman to another—because he has to break up with one and then find the other, which could take months, years, or decades. Viral load will also be lower by then, which makes serial monogamy even safer.


    Illustration of spread of HIV infection with concurrent and serial relationships

    Recent randomised trials have shown that circumcised men are 60-70% less susceptible to HIV than uncircumcised men.16 This has led to the theory that the explosive spread of HIV in the 11 countries along the east and southern arc of the African continent, where national prevalences range from 6% to 24%, is largely attributable to high rates of long term concurrent partnerships and low rates of male circumcision.17

    Concurrency and male circumcision do not explain everything about AIDS in Africa; nor do they imply a simple solution. Nevertheless, behaviour change, especially partner reduction, may be the fastest way to reduce the spread of HIV in this part of the world. Male circumcision is also important, but it is only partially effective in reducing HIV risk, safe affordable circumcision services are still rare in most African countries, and many men may not want to be circumcised. No other intervention to prevent sexual transmission of HIV has proved effective at a population level, including promotion of male and female condoms, probably because condoms are seldom used in the long term relationships in which most HIV transmission in Africa occurs.18

    On the other hand, reductions in sexual partners have been the dominant factor wherever infection rates have fallen in Africa, and the causes have been properly investigated.19 In Uganda, a 60% fall in casual partnerships coincided with a 70% fall in HIV prevalence during the 1990s,20 and similar, though less pronounced, changes in behaviour have also been observed in Zimbabwe and Kenya, where HIV prevalence has fallen more recently.21 The researchers didn’t determine whether people cut back on casual, commercial, or longer term concurrent partnerships, but it may not matter, because in the presence of a “concurrency superhighway” all unprotected sexual encounters are very risky.

    Increased condom use also contributed to these declines, but where condom use alone has increased and partner reduction has not occurred, HIV infection rates have not fallen.22 Universal monogamy is an impossible goal anywhere, but mathematical modelling studies suggest that even small changes in the fraction of people with multiple partnerships can decrease everyone’s risk by breaking up transmission pathways (Morris M, unpublished data).23 When most transmission occurs in long term relationships, only unrealistically high rates of consistent condom use would achieve the same effect. Even the falls in HIV prevalence in the US gay community and Thailand were accompanied by steep drops in multiple partnerships, along with increases in consistent condom use.11 24

    Although strong evidence for the importance of long term concurrency has been available for at least a decade, even today, few, if any, school based AIDS education programmes in Africa warn young people of the dangers that the concurrency superhighway poses to everyone, including those with few, or even just one, trusted long term partner. Instead, most youth prevention programmes continue to stress abstinence, condoms, and youth friendly reproductive health services.25 Thus many young people with only a small number of concurrent relationships, or with a single partner who has another, concurrent relationship, don’t see how risky their behaviour really is.

    Far more research on concurrency is needed, but many technical agencies and African leaders now recognise the urgency of developing programmes to address it. Thus it is disappointing that the UNAIDS secretariat has such a poor record of clarifying the scientific issues surrounding it (box ). This not only slows the fight against AIDS, but raises questions about the agency’s credibility.

    UNAIDS on long term concurrency and behaviour change

    Long term concurrency

    Until 2008, UNAIDS statistical reports on sexual behaviour did not include information on multiple sexual partnerships, let alone long term concurrency. Although UNAIDS surveys did report sex in the past year with a “non-regular” partner, this variable fails to capture people who have multiple long term partners, and who are thus at high risk of infection themselves and of passing the infection to others.

    Until 2006, the public documents of UNAIDS did not mention long term concurrency.26 Then, the agency’s 2008 report on the global AIDS epidemic used a flawed analysis of two epidemiological studies to cast doubt on the concurrency hypothesis, and ignored a large body of evidence in favour of the theory (see bmj.com). UNAIDS still has no best practice document dealing with the risks posed by networks of long term concurrent relationships.

    The agency is currently using mathematical models to characterise the epidemic in various African countries. However, these “Know your Epidemic, Know your Response” studies rely on models that do not account for long term concurrency and make assumptions about sexual behaviour that behavioural studies do not substantiate.27

    Importance of partner reduction

    For years, the agency misrepresented the fall of HIV prevalence in Uganda as being the result of condoms and abstinence,28 even though independent reports showed that partner reduction was the most important behavioural shift.29 30 31 Likewise, UNAIDS has consistently attributed the HIV declines in Thailand and the US gay community to condoms alone, even though partner reduction played a prominent part in these cases too.11 24 The UNAIDS 2008 report also stated that increases in multiple sexual partnerships occurred in countries where infection rates had recently declined,32 even though evidence for such increases does not exist.21 33

    UNAIDS director Peter Piot and colleagues recently warned that addressing multiple concurrent partnerships should not be seen as a “magic bullet” solution.34 They recommend “combination prevention,” the elements of which include campaigns to encourage abstinence, faithfulness, and condom use; male circumcision; voluntary HIV counselling and testing; treatment of other sexually transmitted diseases; research into technologies such as vaccines, microbicides, and pre-exposure prophylaxis with antiretroviral drugs; and programmes to tackle “structural factors” such as gender and economic inequalities that may promote risky behaviour.

    Although all these recommendations are sensible, only male circumcision has been shown consistently to reduce HIV transmission in Africa. Other elements of the combination have had little effect on HIV incidence in the generalised epidemics in east and southern Africa.18 22 35 36 37 One reason may be that none of these programmes provided their target groups with information about the risks of multiple concurrent partnerships, as opposed to unprotected casual sex in general—nor did they deal with the question of why so many African people who don’t engage in typical high risk behaviour contract the virus. In Thailand and in the gay communities of the West, the causes of the explosive spread of HIV were obvious early on, and thus people knew how to protect themselves, and did. In Africa, many people realised too late that their greatest risks came not only from sex workers and casual sex but from the people they are closest to—their spouses and other long term partners.

    Fresh perspective

    The most urgent task facing the new UNAIDS director will be to shift the debate over HIV prevention in a more rational direction, as some within UNAIDS, especially the regional support team for southern Africa, have already done.14 There are legitimate concerns that raising issues of sexual behaviour necessarily implies a moralistic prescription for HIV prevention. However, the concurrency hypothesis in no way implies the superiority, moral or otherwise, of one culture over another—except when it comes to HIV risk. Talking about sex is never easy, and talking about differences in the sexual behaviour of different peoples in different parts of the world is even harder. But with millions of African people dying from AIDS, we had better start doing it.

    The near exclusive emphasis on so called high risk behaviour may be the most destructive misconception about AIDS in Africa. It has probably hindered prevention, promoted denial and stigma, and, by implying that people with HIV are necessarily “promiscuous,” contributed to HIV associated domestic violence.38 It has also promoted the dispiriting sense that there is no rational approach to HIV prevention in Africa, that behaviour change is futile, and that Africans must await further technological wizardry from Western laboratories. This undermines the sense of community ownership that many health officials support. So what policy changes should UNAIDS’s new director make?


    UNAIDS and its governing board should re-evaluate the agency’s political and scientific roles. In particular, scientific issues should be addressed through a more open process of research and peer review, rather than authorised by a single, largely unregulated UN agency.

    UNAIDS should recommend that education about concurrency be integrated into all AIDS programmes in Africa, including those aimed at schoolchildren and young people. Such education should stress that although delay of sexual debut is a sensible goal, personal fidelity is no guarantee of protection against HIV if the partner one eventually ends up with has even one other concurrent partner.

    Behaviour change has been most successful when collective and accompanied by changes in norms and values.39 This is especially true for HIV prevention, because risk of infection is determined not only by personal behaviour but by the behaviour of others. Discussions about concurrency and its implications for sexual relationships could help mobilise broader social change throughout the network, as happened in Uganda and in the US gay community.40 UNAIDS should encourage such discussions through workshops and peer led community meetings, which have already shown promise in improving gender communication and reducing domestic violence.35

    In generalised epidemics, virtually all sexually active people are at risk, and it is this collective recognition of a shared calamity that seems to have been crucial to motivating collective behaviour change among gay men and in Uganda. Therefore, UNAIDS should ensure that the routine testing programmes now expanding across the African continent avoid, as far as possible, creating artificial moral distinctions between people who are and are not HIV positive.

    Sexually transmitted infection services are ideal settings for HIV counselling, but UNAIDS should advise programmes to stop telling people that having their sexually transmitted infection treated will protect them from HIV, since there is scant evidence that this is the case.41

    Finally, the agency should continue to advocate for research into vaccines, microbicides, and other novel prevention technologies, but it should also recognise that much could be achieved without them.

    Further references on concurrency

    • Hudson CP. AIDS in rural Africa: a paradigm for HIV-1 prevention. Int J STD AIDS 1996;7:236-43.

    • Hudson CP. Concurrent partnerships could cause AIDS epidemics. Int J STD AIDS 1993;4:249-53.

    • Hudson CP. Concurrent partnerships and the results of the Ugandan Rakai project. AIDS 1993;7:286-7.

    • Hudson CP. Stable prevalence of HIV-1 in African populations. AIDS 1994;8:1739-40.

    • Morris M, Kretzschmar M. Concurrent partnerships and the spread of HIV. AIDS 1997;11:641-8.

    • Watts CH, May RM. The influence of concurrent partnerships on the dynamics of HIV/AIDS. Math Bios 1992;108: 89-104.

    • Mah T, Halperin DT. Concurrent sexual partnerships and the HIV epidemics in Africa: evidence to move forward. AIDS Behav 2008 Jul 22 [epub ahead of print].

    • Talle A. Desiring difference: risk behavior among young Maasai men. In: Young people at risk: fighting AIDS in northern Tanzania. Copenhagen: Scandinavian University Press, 1995.

    • Epstein H. Why is AIDS worse in Africa? Discover 2004 Feb.

    • Halperin DT, Epstein H. Why is HIV prevalence so severe in southern Africa? The role of multiple concurrent partnerships and lack of male circumcision—implications for AIDS prevention. S Afr J HIV Med 2007;8:19-23.

    • Obbo C. HIV transmission through social and geographical networks in Uganda. Soc Sci Med 1993;36:949-55.

    • Kaleeba N, Ray S. We miss you all: AIDS in the family. Harare: SAFAIDS, 2002.

    • Epstein H, Parkinson S. New wave big enough but no tsunami. AIDSLink 2005 Sep-Oct 2005:93.

    • Mah T. Concurrent sexual partnerships and HIV transmission in Khayelitsha, South Africa. Centre for Social Science Research working paper No 225, August 2008. www.commerce.uct.ac.za/Research_Units/CSSR/Working%20Papers/papers/WP225.pdf.

    • Meyerson B, Robbins A, Koppenhaver T, Fleming D. TCM exposure and HIV related knowledge, attitudes and practices from the 2003 Makgabaneng Listenership survey in Botswana. www.policyresourcegroup.com/pdfs/Botswana_TCM_Evaluation.pdf.

    • Parker W, Connolly C. Namibia: a midterm household analysis of residents from Keetmanshoop, Oshakati, Rundu and Walvis Bay 2007. Windhoek: NawaLife Trust, 2007.

    • Parker W et al. Concurrent sexual partnerships amongst young adults in South Africa: challenges for HIV prevention communication. Johannesburg: Cadre, 2007.

    • Sullivan J, Busang L, Oluput G, Muvandi I. Reported sexual concurrency is a better predictor of HIV status among Batswana females than reported number of sexual partners in the past 12 months. AIDS 2006 16th International AIDS Conference. Abstract No CDC0680. www.iasociety.org/Default.aspx?pageId=11&abstractId=2195103.

    • Colvin M, Abdool Karim SS, Connolly C, Hoosen AA, Ntuli N. HIV infection and asymptomatic sexually transmitted infections in a rural South African community. Int J STD AIDS 1998;9:548-50.

    • Carael M. Sexual behavior. In: Cleland J, Ferry B, eds. Sexual behaviour and aids in the developing world. London: Taylor and Francis, 1995.

    • Standing H, Kisekka MN. Sexual behavior and aids in sub-Saharan Africa—an annotated bibliography. London: Overseas Development Agency, 1989.

    • Carter MW, Kraft JM, Koppenhaver T, Galavotti C, Roels TH, Kilmarx PH, et al. A bull cannot be contained in a single kraal: concurrent sexual partnerships in Botswana. AIDS Behav 2007;11:822-30.

    • Mattson CL, Bailey RC, Agot K, Ndinya-Achola JO, Moses S. A nested case-control study of sexual practices and risk factors for prevalent HIV-1 infection among young men in Kisumu, Kenya. Sex Transm Dis 2007;34:731-6.

    • Lagarde E, Auvert B, Caraël M, Laourou M, Ferry B, Akam E, et al. Concurrent sexual partnerships and HIV prevalence in five urban communities of sub-Saharan Africa. AIDS 2001;15:877-84.

    • Helleringer S, Kohler HP. Sexual network structure and the spread of HIV in Africa: evidence from Likoma Island, Malawi. AIDS 2007;21:2323-32.

    • Baeten JM, Richardson BA, Lavreys L, Rakwar JP, Mandaliya K, Bwayo JJ, Kreiss JK. Female-to-male infectivity of HIV-1 among circumcised and uncircumcised Kenyan men. J Infect Dis 2005;191:546-53.

    • Carael M, Ali T, Cleland J. Nuptiality and risk behaviour in Lusaka and Kampala. Afr J Reprod Health 2001;5:83-9.

    • Voeten HA, Egesah OB, Habbema JD. Sexual behavior is more risky in rural than in urban areas among young women in Nyanza province, Kenya. Sex Transm Dis 2004;31:481-7.

    • Drumright LN, Gorbach PM, Holmes KK. Do people really know their sex partners? Concurrency, knowledge of partner behavior, and sexually transmitted infections within partnerships. Sex Transm Dis 2004 ;31:437-42.

    • Adimora A, Schoenbach VJ, Martinson FE, Donaldson KH, Stancil TR, Fullilove RE, et al. Concurrent partnerships among rural African Americans with recently reported heterosexually transmitted HIV infection. J Acquir Immune Defic Syndr 2003;34:423–9.

    • Adimora A et al. Concurrent sexual partnerships among African Americans in the rural South. Ann Epidemiol 2004;14:155–60.

    • Ankrah M, Asingwiire N, Wangalwa SS, Kyomuhendo S, Misanya-Gessa A. AIDS in Uganda: analysis of the social dimensions of the epidemic. National survey, September–December 1989. Kampala: Makerere University, 1993.

    • Morris M. Concurrent partnerships and syphilis persistence: new thoughts on an old puzzle. Sex Transm Dis 2001;28:504-7.


    Cite this as: BMJ 2008;337:a2638


    • Feature, doi:10.1136/bmj.a2566
    • Competing interests: HE has written a book about AIDS in Africa and consults for organisations working on AIDS prevention.

    • Provenance and peer review: Not commissioned; externally peer reviewed.