Hallett TB, Singh K, Smith JA, White RG, Abu-Raddad LJ, Garnett GP. Understanding the impact of male circumcision interventions on the spread of HIV in southern Africa. PloS One 2008;3:e2212.
To use mathematical modeling to predict the effect of male circumcision on HIV incidence and prevalence in southern Africa
Mathematical modeling using results from clinical trials of the efficacy of male circumcision at reducing incident HIV infections among heterosexual men
Estimates of a) the likely net impact of circumcision interventions; b) the way that this intervention will interact with existing interventions; and c) the potential for perverse impacts of circumcision scale-up programs on HIV transmission
A simple deterministic model was used. Published data from Zimbabwe(1,2) were used to develop sexual behavior parameters. Observational data(3) provided information on disease progression and risk of HIV transmission: primary infection (high infectiousness, short duration), latent infection (low infectiousness, long duration), pre-AIDS (high infectiousness, short duration), and AIDS (death or initiation of antiretroviral therapy [ART], which results in low infectiousness and long duration). The fraction of persons eligible for ART increases from 0%, two years before the circumcision intervention starts (which the authors take to be approximately equal to calendar year 2005) to 28% within 2 years (i.e. the year 2007). ART coverage was assumed to plateau at 90% in 2020. The model permitted individuals to move between groups, thereby holding constant the mean and distribution of risk in the population. Circumcision was estimated to reduce the risk of acquiring HIV by 60%.(4,5) The rate at which men are circumcised in the model is such that the eventual level of coverage of circumcision (fraction of men circumcised) is reached within 5 years of the intervention starting. For some simulations, a healing period of 2 months is assumed. The model population was divided into women, circumcised men, and uncircumcised men.
The model predicts decreases in HIV incidence in circumcised men. There is also an indirect benefit for women and uncircumcised men based upon estimated declines in HIV prevalence following the intervention. As incidence declines among women, circumcised men experience a reduction in risk greater than the 60% attributed to circumcision.
Assuming that the circumcision has no effect on male-to-female transmission and that 50% of men are circumcised, the model predicts a decrease in incidence of 20% after 15-20 years among both men and women. If we assume that circumcision results in a 30% decrease in risk of transmission to women, the effect of circumcision is even greater.
Circumcision coupled with other interventions has an even greater impact than circumcision alone. The greatest effect comes with the combination of circumcision and reduction in risk behaviors. When combined with ART, circumcision results in fewer deaths and infections in both the short and long term.
Circumcision may result in increases in risk behaviors among circumcised men. If circumcised men are 60% less likely to use condoms with casual partners, infections among women will likely increase. However, overall incidence in the population will be reduced by 16%. With a 90% reduction in condom use, infections among women could increase by 40% and the incidence overall would increase.
Resumption of sexual activity prior to complete healing after circumcision may increase transmission. In the model, a very high estimate of 40% of men engaging in sex prior to healing had little effect on new infections among women.
Circumcision will reduce HIV infections but not eliminate them. The reduction in infections is greater when circumcision is combined with reduction in risk behaviors.
The mathematical modeling methods in the study are of high quality. The models were constructed using published estimates of the effectiveness of circumcision in reducing HIV acquisition. Sensitivity analyses were conducted providing a range of estimates based upon changes in parameters. Details on the source of the protective effective of behavioral interventions were not included.
Male circumcision is the HIV prevention intervention with the greatest efficacy. The impact of scale-up programs for male circumcision on HIV transmission is not yet known. The use of mathematical modeling is a practical way of estimating the impact of circumcision under various assumptions. Circumcision was shown to reduce the risk of transmission from women to men by 60%. This study projects a reduction in incidence by 25-30%, with high proportion of men being circumcised. This is comparable to the estimated 37% efficacy of a vaccine provided to both men and women.(6)
A key finding from this study is the prediction that circumcision cannot reduce HIV incidence rates to zero. Combining circumcision with interventions to reduce sexual risk behaviors is likely to have the greatest impact in reducing HIV incidence in southern African. Funding for programs aimed at reducing sexual risk through fewer partners, longer periods of abstinence (such as delaying sexual debut), and condom use should continue.
- Gregson S, Garnett GP, Nyamukapa CA, et al. HIV decline associated with behavior change in eastern Zimbabwe. Science 2006;311:664-6.
- Gregson S, Nyamukapa CA, Garnett GP, et al. Sexual mixing patterns and sex differentials in teenage exposure to HIV infection in rural Zimbabwe. Lancet 2002;359:1896-1903.
- Wawer MJ, Gray RH, Sewankambo NK, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis 2005;191:1403-9.
- Gray RH, Kigozi G, Serwadda D, et al. Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet 2007;369:657-66.
- Bailey RC, Moses S, Parker CB, et al. Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet 2007;369:643-56.
- Williams BG, Lloyd-Smith JO, Gouws E, et al. The potential impact of male circumcision on HIV in Sub-Saharan Africa. PLoS Med 2006;3:e262.