Helen Epstein

Concurrent sexual partnerships help to explain Africa's high HIV prevalence: implications for prevention

As Kiat Ruxrungtham and colleagues describe in today’s Lancet HIV transmission in most Asian countries remains strongly associated with particularly high-risk activities—ie injection-drug use male-male sex prostitution and in China paid donation of plasma. Although there is understandable concern that the virus could soon spread widely through the general population. HIV has been present in Asia for nearly two decades and such extensive spread has yet to occur. For example analysis of trends in India suggests that HIV prevalence both in high-risk groups and in the generally low-risk antenatal clinic population has probably stabilised in recent years. It is possible that large-scale heterosexual epidemics will never emerge in most of Asia except perhaps on the island of Papua. Furthermore in some of the world’s most populated countries—Pakistan Bangladesh Indonesia and the Philippines home to some one billion people-nearly all men are circumcised further restricting the potential for extensive heterosexual spread. In chilling contrast as Emil Asamoah-Odei and colleagues report also in today’s Lancet HIV rates remain very high in much of east and especially southern Africa. The overwhelming burden of HIV/AIDS is still concentrated in this region which accounts for only 3% of the global population yet some 50% of global HIV cases. For example infection rates in adults in South Africa Botswana Zimbabwe and western Kenya range from 20 to 40% roughly an order of magnitude higher than anywhere else in the world. (excerpt)

Measuring concurrent partnerships.

We are encouraged that UNAIDS is developing new indicators to measure concurrent sexual partnerships (Feb 20, p 621). However, we believe that adding a measure of coital frequency to the agency’s proposed list of questions would provide an important improvement. Each additional act of intercourse is associated with a signifi cantly increased risk of infection and thus coital frequency is crucial to the risk of HIV acquisition. Consider a scenario in which a man with two partners has sex with one of them only a few times a year and with the other a dozen times a year. This man would be regarded as practising concurrency according to the UNAIDS defi nition, but even if such behaviour were universal among men and women in a population, it would be very unlikely to generate a signifi cant AIDS epidemic. New modelling research suggests that sustained heterosexual HIV transmission requires that a signifi cant number of overlaps be long enough and coitally frequent enough so that a relatively large number of people have sex with someone who also has another partner, particularly while some of these individuals are in the highly risky “acute phase” of early HIV infection. Little research on coital frequency in concurrent partnerships has been done, but some studies suggest that it may be quite high in some high-HIVprevalence hetero sexual populations that practise concurrency—even when overall partner numbers are low. We recommend that, in addition to the questions listed by UNAIDS, an additional one such as the following be asked for each overlapping partner: “Let’s talk about this partner. During the past year, how often do/did you have intercourse with him/her? (a) only once; (b) more than once, but less than once per month; (c) about once per month; (d) a few times per month; (e) about once per week; (f) about two or three times per week; (g) more than two or three times per week.” Furthermore, we recommend that UNAIDS and others explore better ways to collect more accurate selfreported data on intimate sexual behaviour. Under-reporting of risky sexual activity, especially by women, is a signifi cant problem, particularly if strict confi dentiality is not perceived by inter viewees.

The myth of the virgin rape myth

www.thelancet.com Vol 374 October 24, 2009 1419 1 Samarasekera U. Countries race to contain resistance to key antimalarial. Lancet 2009; 374: 277–80. 2 Yeung S, Van Damme W, Socheat D, White NJ, Mills A. Access to artemisinin combination therapy for malaria in remote areas of Cambodia. Malar J 2008; 7: 96. 3 Denis MB, Davis TM, Hewitt S, et al. Effi cacy and safety of dihydroartemisinin-piperaquine (Artekin) in Cambodian children and adults with uncomplicated falciparum malaria. Clin Infect Dis 2002; 35: 1469–76. 4 Janssens B, van Herp M, Goubert L, et al. A randomized open study to assess the effi cacy and tolerability of dihydroartemisininpiperaquine for the treatment of uncomplicated falciparum malaria in Cambodia. Trop Med Int Health 2007; 12: 251–59. 5 Tran TH, Dolecek C, Pham PM, et al. Dihydroartemisinin-piperaquine against multidrug-resistant Plasmodium falciparum malaria in Vietnam: randomised clinical trial. Lancet 2004; 363: 18–22.

HPTN 052 and the future of HIV treatment and prevention.

The findings of the HIV Treatment for Prevention Trial HPTN 052—ie, a 96% reduction in HIV transmission within discordant couples1—confirms that early treatment is a potent intervention for clearly defined couples. But its implementation on a population scale might not be so successful, and we are concerned about calls such as that in your May 21 Editorial2 to dispense with behavioural HIV prevention pro grammes in favour of this approach. In high-prevalence African countries, where half of all new infections globally now occur, early treatment will not prevent the roughly 40% of infections estimated to occur during acute infection,3 or the 30–60% of new infections in “stable” couples that originate outside the couple.4 Indeed, it did not prevent the seven to 11 infections (of a total of 40) that seem to have originated from outside the couples in the HPTN 052 trial.1 Of course, this risk would be eliminated if absolutely every HIV-positive person were treated, but this seems a utopian goal, given the state of African health systems. In reality, attempts at mass treatment would almost certainly accelerate drug resistance, increasing the already overwhelming costs and logistical challenges of treating those who need it most urgently. Behavioural prevention has saved millions of lives around the world, the best documented cases being the national campaigns of Uganda and Thailand and the internally designed awareness-raising within the gay communities of western countries during the 1980s and early 1990s. When behaviour-change programmes succeed, they tend to involve collective changes in norms and behaviour, preceded by the development of some sort of community consensus.5 International development agencies’ pro grammes could be greatly improved if they informed people about the dangers of long-term concurrent sexual partner ships—in addition to casual sex4—and worked more closely with people at risk to develop their own responses to this threat. The Tostan programme has used such an approach to reduce the practice of female genital mutilation, and others could used it to fight HIV as well.

HIV vaccine trials for Uganda

The high incidence of HIV infection and stability of the government over the past 8 years make Uganda an attractive place for HIV vaccine trials. But is the population open to the idea of participating? At a conference last month entitled Preparation for HIV Vaccine Evaluations in Uganda organized by the Ugandan government the US National Institutes of Health Case Western Reserve University and WHO [World Health Organization] registrants heard that in a survey conducted by Makerere University and University of California San Francisco 95% of 201 seronegative STD clinic attendees said that they would be willing to participate in an HIV vaccine trial. But 20% of these people said that they would use condoms less often if they were in a trial. The finding underscores the importance of making subjects understand the nature of a randomized placebo-controlled trial. (full text)