Male Circumcision, In So Many Words…

Interest in male circumcision as an HIV prevention intervention is focused on the promise it holds to reduce the risk of HIV infection for men engaged in heterosexual, vaginal intercourse. Following an international consultation on 6–8 March 2007, WHO and UNAIDS released “Conclusions and Recommendations” on the policy and programmatic implications of the existing data on male circumcision and HIV prevention.¹ This is a welcome step in what will inevitably be a long process to ensure male circumcision is appropriately implemented as part of the response to HIV. Beyond simply the offer of the service, if male circumcision is to be an effective strategy over the long term, these conclusions and recommendations will need to be translated into guidelines and adapted and adopted at national and local level. Policies and programmes will need to be developed, and appropriate monitoring and evaluation systems created. In doing so, the devil, as they say, will be in the details. Some issues to be considered as this process moves forward are outlined below.

Words matter

Often the careful nuances put forward in global policy statements have been lost when translated into programmatic efforts and even more so in how the concepts have been understood by those who have received the intervention – and in this case also by those with whom they have sex. What are the actual sexual acts that “reduce the risk of heterosexually acquired HIV infection in men”? The language is vague and insufficiently clear. If, as the current evidence suggests, it means that circumcision is partially protective only of the man who performs vaginal penetrative sex, then this will need to be stated clearly at every turn. The implications for women who are recipients of vaginal penetrative sex with circumcised men are unknown. Also unknown are the risks of exposure for the receptive female or male partner of a circumcised man during anal sex and for the circumcised male performing penetrative anal sex. The implications of these unknowns for the success of the intervention at both individual and population levels will need to be carefully considered.

With further regard to semantic clarity, no definition is put forward by WHO and UNAIDS as to what exactly constitutes male circumcision: how much of the foreskin has to be removed in order for the protective effect to be conferred upon a male? Traditional male circumcision is practised by many different communities, with a wide degree of variability in terms of how much of the foreskin is removed (sometimes as little as 1–2 cm is removed compared with the 4 cm removed during most medical circumcision).² The potential exists that men circumcised through traditional practices might not benefit from the partial protection from HIV acquisition during vaginal sex conferred through removal of more of the foreskin. This too will need to be clarified in any further policy and programmatic guidance offered.

Public health messaging at all levels will have to ensure that it is clearly understood by all concerned, now and over time, that circumcision is only known to be partially protective for the circumcised man engaging in penetrative vaginal sex. Male circumcision has to be presented as “an important intervention to reduce the risk of HIV acquisition by men through vaginal sex” and not simply an “an important intervention to reduce the risk of heterosexually acquired HIV infection”*. All health messages, in whatever ways they are designed to be appropriate to local cultures and contexts, will need to be absolutely explicit about the relative risks and potential benefits for each partner engaging in vaginal or anal sex.

When are adult men who are circumcised sufficiently healed to have sex?

The WHO/UNAIDS document recommends that men who undergo circumcision abstain from all sexual activity for at least six weeks following the operation.¹ This guidance is based on data from the three randomised clinical trials and preliminary data reported in a press release during the consultation, suggesting that HIV positive men who have unprotected sex before complete wound healing may have a greater risk of wound complications and be more likely to transmit the virus to their female partner (no research exists yet with regard to men who have sex with men in this context).³ Moreover, evidence from the three clinical trials shows that 11–14% of men had sex before their circumcision wounds were totally healed, despite intensive counselling against any sexual intercourse during this period.^4–6 Outside a clinical trial setting, where the emphasis on counselling will inevitably be far less, it is likely that the rates of sexual activity for men before their wounds heal completely could be far higher.

Must we keep repeating history? This is reminiscent of the spread of genital warts postulated to have been associated with increased rates of HIV infection among men having sex with men in San Francisco and other US cities in the early 1980s.^† There, men who were having their warts removed were not allowing sufficient time for the wounds to heal before having sex again, and here men are having their foreskins removed and it appears engaging in unprotected sex before they are fully healed. The health implications if men fail to wait the necessary period post-circumcision before penetrating their sexual partners, both for themselves and their partners, are as yet unknown, but the lesson from these early days of HIV needs to be remembered. In places where a return to the health centre for a determination that the man is now free to engage in sexual activity is not feasible, innovative health messaging is needed that can be rigorously tested for its ability to promote the required period of abstinence and recognition by men of complete healing, in addition to condom use, when resuming sex.

Power relations, not just sexual relations

As perceived risk of HIV infection decreases for circumcised men, what effect might this have on their willingness to use condoms? Much has been made in recent years of the relationship between power and the ability to negotiate condom use, with the power traditionally assumed to be with the penetrating partner.⁷ A number of studies have also tried to understand the relationship between the perception individuals have of their own risk and vulnerability to HIV infection and their use of condoms during sexual activity.⁸ Despite the paucity of studies and the consequent inadequate understanding of these relationships, one thing is clear: the perceptions that people (including both women and men) have of their risk impacts on their use of condoms even if they are available.⁹ Male circumcision having been found to be protective of the penetrating partner during vaginal sex, the potential decline in that partner's risk perception has clear implications for the receptive partner's ability to negotiate condom use.

None of the three randomised clinical trials found evidence of “disinhibition” among circumcised men (i.e. higher risk behaviour due to feeling more protected by circumcision), and it is important to note that, in Rakai, the condom use rate increased from 14% to 19% in both arms of the trial.¹⁰ Even as the content of this counselling has not yet been made public, one can assume this is an illustration of the very positive impact that intensive counselling can have on behaviour. Yet, outside a clinical trial setting, where intensive counselling is likely not provided to the same extent, the potential impact on condom use is insufficiently understood and of deep concern. In addition to counselling constituting an absolutely vital component of any male circumcision service package, messages are required for the broader population to promote risk awareness and prevention not only for circumcised men but for their sexual partners as well.

Some will argue that men who seek circumcision may not use condoms irrespective of their circumcision status and, if circumcised, at least they would benefit from partial protection. However, with women already disproportionately vulnerable to and burdened by HIV, it is critical that gender dynamics and the potential negative effects on women of this new HIV prevention intervention not be simply noted (or addressed as an after-thought) but form the cornerstone of any newly framed HIV prevention strategy.

What is needed for the “protective effect” of circumcision to be felt at the population level?

The local prevalence of male circumcision is assumed to play an important role in determining the potential for male circumcision to lower the risk of HIV acquisition at a population level. WHO indicates that it will take many years for the population-level impact of male circumcision as an HIV prevention tool to be seen.¹¹ Actual risk reduction at population level is thought to require an initially low rate of male circumcision to be increased to 50–80%.¹² Until then, individual men who are circumcised will be more protected during vaginal sex, but without condom use their partners will not.

The needed rise in prevalence of circumcision and the sustainability of its benefits over time would therefore appear to require that programmatic attention be given to infant, early childhood and pre-pubertal male circumcision. This has important implications where the uptake of male circumcision may, for social, cultural or religious reasons, never reach these high levels in adults. Importantly, it also raises the question of how governments will determine the “best interest of the child” and in what ways this will be defined over time.

Striving for population level protective effects must go hand in hand with effective monitoring and evaluation of circumcision interventions. This requires not only focusing on the health sector and on how the surgery is performed, but in order to ensure that the current push towards circumcision succeeds over time to documentation of successful approaches to addressing larger legal, policy and community barriers.

The search for a quick fix

Much of the current debate, as well as the recommendations from WHO and UNAIDS, have focused on the potential value of male circumcision for protecting men engaged in vaginal sex in developing countries. Yet, in the US, both the US Centers for Disease Control and Prevention and the New York Health Department are already investigating the possibility of adopting policies to encourage male circumcision.¹³¹⁴ In New York City, the suggested focus is on men considered to be at highest risk of infection, which health officials understand to include men who have sex with men, African American men and men attending clinics for sexually transmitted infections (with no specificity at this point as regards whom the latter two groups are having sex with nor what sorts of sexual acts they engage in). Nor do they seem to be considering that no reduction in transmission risk has yet been shown in anal sex between two men. It is well established that the efficiency of sexual transmission of HIV is highest for receptive anal sex, and in descending order for receptive vaginal sex, penetrative anal sex and penetrative vaginal sex.¹⁵ Messages on the impact that circumcision may have on individual risk reduction, depending on sexual practices, must be clearly articulated before programmes are rolled out. Preliminary data indicate that demand for male circumcision is already increasing in some places,² suggesting that the need for disseminating accurate information is pressing.

Solid research has, over almost two decades, succeeded in putting male circumcision on the HIV prevention agenda. While it would have been unethical not to move forward with the accumulated evidence, the need to “say it how it is” cannot be overstated. Shying away from language that is sexually explicit and descriptive leaves space for misinterpretation and misapplication in health policy and programmatic guidelines with potentially devastating impacts. Furthermore, the necessity to effectively monitor and evaluate the practicality and impact of all components of male circumcision interventions, both individually and overall, is critical for both sound public health and human rights. The prevention agenda now being contemplated needs clarity, simplicity and close monitoring if it is to be effective and safe for both men and women who engage in sexual activity.

Notes

* These distinctions may seem small but have tremendous importance for whether men understand themselves to be protected through male circumcision, with particularly worrisome impacts for women who may think themselves protected if the focus on men is not sufficiently emphasised.

† Personal communication with friends and public health colleagues, San Francisco, CA, April 2007.