Rapid Responses to:

EDITORIALS: Quarraisha Abdool Karim Prevention of HIV by male circumcision BMJ 2007; 335: 4-5 [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

On the potential impacts of male circumcision on Africa’s HIV epidemics from modelling and simulation

Cyrille Delpierre, Lise Cuzin, Jeffrey B. Bingenheimer   (20 September 2007)

On the potential impacts of male circumcision on Africa’s HIV epidemics from modelling and simulation 20 September 2007
Cyrille Delpierre, Ph.D Inserm U558 Toulouse, 31000, France, Lise Cuzin, Jeffrey B. Bingenheimer Send response to journal: Re: On the potential impacts of male circumcision on Africa’s HIV epidemics from modelling and simulation	On the potential impacts of male circumcision on Africa’s HIV epidemics from modelling and simulation About male circumcision, the main question remains: What effects could the scale-up of male circumcision [MC] have on the population-level dynamics of the HIV epidemics in sub-Saharan Africa? Efforts to answer this question have taken the form of mathematical modelling and simulation exercises 1-4, inevitably involving considerable uncertainty. Karim et al. 5 identified surgery as a key point of uncertainty. Assuring the same quality than in trials, by using trained medical team for surgery and counselling, with the same standard of care, particularly the respect of healing period, are central points for implementing with success circumcision program. Although we agree that the way to integrate circumcision program in existing sexual health services may be “the biggest challenge”, other points of uncertainty need to be raised once the surgery is done. The potential behavioral impacts of MC constitutes another significant source of uncertainty. Now the susceptibility-reducing effect of MC is quickly becoming common knowledge in many parts of sub-Saharan Africa, men are likely to receive a very different message about the protective effects of MC. Under these circumstances, circumcised men may decrease their condom use or engage in riskier behaviour because they perceive that MC provides full or partial protection against HIV. Another reason to expect some impact of MC on sexual behavior involves changes in the balance of power between women and men. MC could make it more difficult for women to bargain for condom use with their male partners. Circumcised men will be in a stronger position to argue that it is unnecessary for them to use condoms since they have already taken dramatic steps to reduce their risk of infection. The uncertainty here involves the nature and magnitude of these behavioral impacts, and the population-level consequences thereof. Gray and colleagues 1 found that a 25% increase in the number of sexual partners could completely wipe out the population- level effects of MC, and that a larger increase could cause HIV epidemics to become substantially more severe than the baseline scenario. Another consideration with respect to existing simulation models involves uncertainty in estimates of current MC and HIV prevalence levels. The prevalence of MC used in some simulation models is based upon data that are several decades old and therefore must be considered with caution. Moreover, using an average level of prevalence for a specific country does not take account of the diversity that exists according to ethnicities and geographical regions. Similarly, for many countries in sub -Saharan Africa, historical and current estimates of HIV prevalence may be inaccurate. This is especially true for countries that have relied upon a small number of sentinel surveillance sites (e.g., antenatal clinics attendees in urban areas) to obtain estimates of national HIV prevalence. For many countries, estimates have improved in recent years as a result of more systematic efforts to estimate national HIV prevalence levels, but this progress has not been uniform. Clearly, inaccurate figures on current levels of MC or HIV prevalence may lead to an incorrect baseline and misleading conclusions for simulation studies. Finally, leaving aside the simulation models, country level estimates of current MC and HIV prevalence provide a sobering reminder that other powerful forces besides MC are at work, such as social, cultural, historical, religious or sexual factors. Consider for example Rwanda and Zimbabwe in which it is estimated than 10% of men are circumcised 4. In Rwanda, HIV prevalence is only 5% and in Zimbabwe it is 25% 4. These figures strongly suggest that the association between MC prevalence and the course of HIV epidemics is far from perfect. MC may be an important tool for reducing the spread of HIV in some parts of sub-Saharan Africa. But we must not allow our enthusiasm for this intervention to prevent us from continuing to develop the fullest possible understanding of the numerous social and biological forces that interact to fuel the spread of HIV in sub-Saharan Africa and beyond. At the individual level, MC provides only partial protection against HIV infection. At the population level, existing simulation models, with all of their limitations, suggest that widespread MC is unlikely to provide complete control of HIV epidemics. REFERENCES 1. Gray RH, Li X, Kigozi G, Serwadda D, Nalugoda F, Watya S, et al. The impact of male circumcision on HIV incidence and cost per infection prevented: a stochastic simulation model from Rakai, Uganda. Aids 2007;21(7):845-50. 2. Kahn JG, Marseille E, Auvert B. Cost-effectiveness of male circumcision for HIV prevention in a South African setting. PLoS Med 2006;3(12):e517. 3. Nagelkerke NJ, Moses S, de Vlas SJ, Bailey RC. Modelling the public health impact of male circumcision for HIV prevention in high prevalence areas in Africa. BMC Infect Dis 2007;7:16. 4. Williams BG, Lloyd-Smith JO, Gouws E, Hankins C, Getz WM, Hargrove J, et al. The potential impact of male circumcision on HIV in Sub-Saharan Africa. PLoS Med 2006;3(7):e262. 5. Karim Q. Prevention of HIV by male circumcision. BMJ 2007;335:4-5. Competing interests: None declared