This literature review follows the logical and steady progress over 20 years of studies and research that led, in 2007, to the WHO and UNAIDS recommending MMC as an essential element of a comprehensive HIV prevention package. Academic and programme modelling over the following four years led to PEPFAR setting targets and determining the required financial support for 13 countries in eastern and southern Africa to scale-up MC programmes to reach and then maintain an 80% MC prevalence. Data released on March 2012 show that these countries have not yet been able to reach the necessary scale-up volumes. The review concludes with the recent WHO announcement of an innovative non-surgical device method, PrePex, which could potentially have a significant effect on enabling these countries to reach these scale-up targets more quickly.

Global Scale of HIV/AIDS Epidemic

At the end of 2011, 34 million people were living with HIV/AIDS (PLHA). 23.5m of these are living in sub-Saharan Africa (SSA), meaning that this region, which has just 10% of the world’s population, bears more than two thirds of the burden of this disease. In recent years there has been an increasing focus on providing treatment for those infected. In 2011, 8m PLHA in low and middle income countries were receiving necessary antiretroviral therapy (ART), an increase of 1.4m from 2010 (UNAIDS, 2012).

In SSA in 2003 only 100,000 received ART; in 2011 this had reached 6.2m, largely funded by the United States President’s Emergency Plan for AIDS Relief (PEPFAR). Whilst this reflects very significant progress, it represents just 56% of those that need treatment. The Joint United Nations Programme on HIV/AIDS (UNAIDS) and the World Health Organisation (WHO) have set a target of reaching 15m on ART by the end of 2015 (UNAIDS, 2012). This is an admirable vision but one that comes with a significant cost. The estimated lifetime cost of ART, in SSA, is $7,400, (Bollinger, DeCormier Plosky and Stover, 2009), so reaching 15m patients will raise the total discounted cost to $111 billion.

The following data give a more complete view of how this epidemic is progressing. During 2011, 1.7m PLHA died from an AIDS related illness, 1.2m of these in SSA. In the same period 2.2m became infected, 1.5m in SSA. More became infected than those either starting treatment or dying from the illness, which means the epidemic is still growing. The number living with the infection continues to increase by one every 56 seconds, more than 560,000 in this next year (UNAIDS, 2012). Even if only half of these newly infected are treated, the lifetime cost will be an additional $8 billion.

HIV/AIDS is very significant global health issue. Notwithstanding all of the worthy campaigns encouraging prevention through behaviour change and the use of condoms, it is not working; people are still becoming infected and in large numbers. Of the 34m PLHA, about 90% are adults above the age of 15 and in SSA most of those becoming infected do so as a result of heterosexual transmission (UNAIDS, 2012).

Early Epidemiological and Observational Studies

HIV prevalence shows a high degree of heterogeneity; notwithstanding similar risk factors, infection rates vary widely between, and sometimes within, different countries. In the 1980s the infection rate was highest in SSA and varied significantly from one country to another and from one population group to another. By the late 1980s some researchers had started to notice, from observational and epidemiological studies, that infection rates were often lower among those groups that also had a high prevalence of male circumcision. In 1986, the first paper suggesting that MC might offer protection against HIV infection was published in the New England Journal of Medicine (Fink, 1986).

A prospective study in 1989 followed 422 men who had acquired an STI from a particular group of prostitutes in which HIV prevalence was 85%. This study investigated the differing rates of seroconversion for those with one or more of three risk factors: frequent sexual contact with these prostitutes, presence of genital ulcers and being uncircumcised. It concluded that the data showed a causal relation between genital ulcer disease (GUD), lack of circumcision and an increased risk of becoming HIV infected (Cameron et al., 1989).

In 1990 a study reviewed the practice of male circumcision, and measured HIV seroprevalence, in 700 population groups in 140 different locations across 41 African countries. Its findings concluded that lack of MC is associated with an increased risk of HIV infection (Moses et al., 1990).

A review in 1994, of 30 epidemiological studies that had previously investigated the possible link between MC and HIV infection, concluded that the then current body of evidence was showing MC to be a causal factor in reduction of HIV transmission. The authors concluded that this evidence met at least 6 of Hill’s 9 criteria for causal inference (Hill, 1965). It was not certain that MC would meet the criteria Specificity as there could be some confounding with STIs. The criteria Biologic Gradient was deemed not to be applicable as it usually referred to dose-effect and MC is one-time, all or nothing. In their discussion they called for further prospective observational studies, randomised controlled trials (RCTs) and studies into the feasibility of using MC as an intervention for improved prevention of HIV infection. These RCTs would meet the one remaining criterion, the need for experimental evidence (Moses et al., 1994).

In 1999 the Lancet published an essay by Halperin and Bailey in which they called time on the debate. They stated that in the 10 years since Cameron’s prospective study there had been another 17 studies in 8 different countries, all of which supported MC as contributing partial protection against female to male HIV transmission (Halperin & Bailey, 1999).

This was quickly followed by a systematic meta-analysis, reviewing 27 observational studies. This review concluded that uncircumcised men were at least twice as likely to become HIV infected compared to those that were circumcised. The rate of protection was even higher for those men considered to be in high risk groups such as truck drivers, those with STIs and in particular those with GUD. The authors noted that an inherent weakness in observational studies is the inability to adequately control for confounders and so recommended RCTs that could provide empirical data to validate these observational findings (Weiss et al., 2000).

Biological Reasons why MC Might Prevent HIV Infection

There are widely accepted biological reasons by which circumcision might provide this degree of protection. Small tears and breaks in the foreskin or trauma to the non-keratinised inner mucosal surface of the foreskin, often as a result of intercourse, can lead to an increased vulnerability to infection. Removal of the foreskin and keratinisation of the glans after circumcision protects against such trauma. The foreskin provides an ideal environment for increased infection with GUDs and other STIs, which are known cofactors for HIV infection (Halperin & Bailey, 1999; Weiss et al., 2000). Cameron and his colleagues noted that the presence of GUDs greatly amplified the risk of becoming infected with HIV and that GUDs were more common, and thus more likely to lead to HIV infection, amongst uncircumcised men (Cameron et al., 1989).

From immunohistochemistry investigations it has been discovered that the inner mucosal surface of the foreskin has a higher density of Langerhans cells nine times that of the skin of the vagina, cervix or rectum. This part of the foreskin is exposed when the penis is erect and the outer foreskin is pushed back during intercourse. These cells are primary targets for the sexual transmission of HIV and whilst also present on the outer skin, and that of the glans, there they are protected by a layer of squamous epithelial cells (Bailey et al., 2007).

Randomised Controlled Trials

These observational studies had formed the hypothesis that men who are circumcised are less likely to become infected with HIV, that somehow circumcision provides a protection rate of about 50-60%. The most recent studies called for this theory to now be tested by experimental trials.

The first such randomised controlled trial (RCT) was conducted at Orange Farm, close to Johannesburg in South Africa, with participant recruitment starting in July 2002; 3,274 men were recruited from the general population. These were divided randomly into one of two groups; an intervention group in which all were immediately offered circumcision and a control group in which participants were asked to wait until the end of the trial, at which time circumcision would be offered. At predetermined times during the trial the participants returned for follow-up during which different information was gathered regarding sexual behaviour, frequency of intercourse, number and status of partners, use of condoms etc.. Participants received an HIV test and were examined for STIs at each follow-up. The trial was stopped early by the data and safety monitoring board after the intermediate analysis, using data gathered up to November 2004, showed conclusively that MC was protective; it was then considered unethical to withhold the intervention from the control group. The data at that time indicated that MC provides a 63% protection rate against HIV infection, a 0.37 relative risk of becoming HIV infected, which validated similar findings from the observational studies. The study concluded that the degree of protection offered by MC is high and consequently it should be included in national plans for the prevention of HIV. Further research and modelling was recommended to assess the potential impact that such an intervention might have (Auvert et al., 2005).

When the results from the Orange Farm trial were announced in July 2005, the WHO issued an interim statement indicating that whilst evidence was building to support the use of MC as part of an HIV prevention programme, this should not be implemented until the results from 2 other, on-going, RCTs became known. The second RCT was performed in Kisumu, Kenya, and started recruitment in February 2002. A total of 2,784 men were randomised into 2 groups in a manner similar to Orange Farm. This trial was stopped early by the data and safety monitoring board after a third interim analysis of data in December 2006. By then the mean time for follow-up was 24 months and the RR, after adjusting for non-adherence, was 0.40 (Bailey et al., 2007).

The third, and final RCT, with 4,996 men in Rakai, Uganda, which started in August 2002, was also stopped early by the data and safety monitoring board in December 2006. The final results from this trial matched that in Kenya, indicating a 60% protective rate. This study showed that the HIV infection rate was lower in the intervention group, compared to those in the control group, across all sub-groups determined either by sociodemographic, behaviour or those with STIs (Gray et al., 2007).

WHO and UNAIDS Recommend MC as Prevention

In March 2007, after an international conference in Montreux, Switzerland which was attended by a wide range of experts and global stakeholders, the WHO and UNAIDS issued a set of recommendations regarding the deployment of male circumcision. It referred to the overwhelming evidence from the three RCTs, how these supported the findings from the observational studies and were consistent with biological investigations. Male circumcision was now being recommended and endorsed by the WHO as an integral part of comprehensive HIV prevention programmes. It strongly recommended countries with high HIV prevalence rates, infection spreading predominately through heterosexual transmission and with a low prevalence of MC, to start scaling up MC services and to especially prioritise this among the high risk groups. This set of recommendations was to become pivotal in the ensuing development of male circumcision programmes across east and southern Africa. It recognised that MC could have a significant impact on the reduction of HIV infection rates, but only if certain issues were addressed. These included concerns about the potential impact on health systems that were already overwhelmed, lacking sufficient human and financial resources. MC was to be included as part of a comprehensive prevention package; protection is only partial and it would be important that circumcised men did not start behaving as if they were now totally immune. Further research would be required to help establish optimal community engagement, operational and scale-up models (WHO and UNAIDS, 2007).

Gaining WHO endorsement of MC as an essential intervention had taken about 20 years, it first being suggested by Fink in 1986. MC is a simple procedure with few side effects and it can be delivered by trained staff with small numbers of adverse events (AE), most of which are easily reversed. The AE rates in the RCTs varied from about 6% in Orange Farm to less than 2% in Kisumu. This lower rate was noted as possibly being due to the procedure being performed by internal, highly trained and experienced physicians, whilst in Orange Farm the procedure had been outsourced to local GPs (Auvert et al., 2005; Bailey et al., 2007; Gray et al., 2007). Research now confirmed that MC provides a 60% protection against female to male transmission of HIV. This one-time procedure offers protection that is thought to be lifelong, which is very different from other prevention measures, such as condom use, PMTCT or treatment as prevention which demand consistent behaviour and adherence. In March 2007, there were many who may have considered this WHO endorsement to be a turning point in the fight against AIDS. Viewed from a 2012 perspective, one can now see that whilst pivotal, this recommendation was just the first of many necessary steps, some of which are still to be taken. There was now formal acceptance of male circumcision as an innovative prevention measure, but how this could be applied to national HIV/AIDS campaigns, what it would cost or what the impact might be, were all unknown.

Cost-effectiveness Studies

At the end of 2005, a paper was submitted concluding that if the results from Orange Farm could be extrapolated to other places and other population groups, and if full coverage of MC could be achieved within the next 10 years, then almost 6m new infections could be averted over a 20 year period, reducing AIDS related deaths by 3m. Future, more detailed, modelling by Njeuhmeli et al. in 2011 would project numbers that are about half of these. The authors noted the need to wait for the results from the other 2 RCTs before confirming this analysis and suggested the need for a study into the cost-effectiveness of MC as prevention (Williams et al., 2006).

The first cost-effectiveness study into the use of MC as prevention against HIV infection was completed in March 2006 and published in December of that year. The analysis was based solely on the results at Orange Farm and considered the impact of circumcising a hypothetical cohort of 1,000 men. The actual results are not so relevant now, since the more detailed and specific modelling by Njeuhmeli et al., but the method is relevant. Costs included those for community engagement and mobilisation, performing the circumcision procedure and management of any resulting AEs. This study suggested that cost-effectiveness should be measured by estimating the number of HIV infections averted (HIA) by MC and to then derive a discounted value for the future costs of treatment that were avoided. Sensitivity analysis was used to model the impact of variance in costs, MC coverage, AE rates and the protective effect. The authors concluded that MC is a cost-effective prevention method and that its cost per HIA compares favourably with other prevention methods such as distribution of condoms, Voluntary Counselling and Testing (VCT) and Prevention of Mother To Child Transmission (PMTCT) (Kahn et al., 2006). At the end of 2006 the relationship between the number of circumcisions and the number of infections averted was still not known. Establishing these data will be critical to the research and future modelling.

The two studies above are referenced by WHO and UNAIDS in the March 2007 report on conclusions and recommendations. Whilst noting these early conclusions that MC is a cost-effective means of prevention, it recommended that each country should now create its own specific, detailed, costed plans for expansion of MC services (WHO and UNAIDS, 2007). It would take another 4 years before this could be started. Programme planners still needed to determine the costs involved in scale-up and importantly there was still no accepted metric showing how many circumcisions would be needed to gain one HIA. Given the multiple factors that influence incidence, this metric would surely vary from one country to another, and perhaps even from one population group within a country to another? Knowing the value for this metric would be critical to the development of national plans.

An expert group was formed by WHO, UNAIDS and the South African Centre for Epidemiological Modelling and Analysis (SACEMA). In 2008 this group reviewed 6 academic mathematical models for the cost-effectiveness of MC on HIV prevention. In a paper published in 2009 the group concluded that even when multiple factors were varied in each of these models, the results consistently indicated that MC would be cost-effective and have a significant impact on HIV incidence and prevalence at a population level. The group noted the recent development by the Futures Institute, in collaboration with USAID Health Policy Initiative, of the Decision Makers’ Program Planning Tool (DMPPT) (Bollinger, DeCormier Plosky and Stover, 2009). The group concluded that since the published results from academic modelling could not easily be extrapolated or applied to country specific situations, the DMPPT would enable country teams to perform the necessary country-specific modelling and the tool would produce results that were consistent with the earlier academic models (UNAIDS, WHO and SACEMA, 2009).

In the years leading up to the end of 2009, WHO, UNAIDS and other agencies had produced a portfolio of documents and guidelines that could be used by country teams to help plan and implement successful scaled-up programmes. These covered a wide range of topics including the technical aspects of training and delivering circumcision surgery, key policy issues, legal matters related to human rights and a communications framework. The availability of the DMPPT now completed the necessary suite of tools to enable country-specific scale-up plans to be developed.

Country Scale-Up Modelling Using DMPPT in 2011

In 2011 Njeuhmeli, Forsythe and colleagues completed a modelling exercise using DMPPT and published findings indicating detailed cost, and impact, of expanding MC services in 13 countries across eastern and southern Africa. They concluded that reaching 80% MC prevalence across this region by 2015 and maintaining that level through to 2025 would avert more than 3.3m new HIV infections. This would require 20.3m circumcisions up to the end of 2015 (activity in the period 2011 – 2015 being referred to as catch-up) and a further 8.4m in the following ten years. Total costs to do so would be about $2 billion but would deliver more than $16.5b in savings from averted care and treatment costs. This modelling indicated that by 2025 half of the infections being averted would be in women, thus confirming the 2009 prediction, by the UNAIDS – WHO – SACEMA expert group, that MC would have a population wide impact on HIV incidence and prevalence (Njeuhmeli et al., 2011).

There are many published works detailing how to establish the unit cost of providing a particular service. These recommend that an ingredient, bottom-up, approach is first used to determine each of the applicable direct costs, specific to the service being provided, such as staff and supplies. Then a step-down approach is needed to apportion indirect, shared costs, such as those for administration, HR and Finance, and use of facilities and equipment (Shepard et al., 1998; Odaga & Okiria, 2010). The DMPPT has been designed to accomplish this and leads a user through the process of entering the required data to generate a unit cost. Njeuhmeli and colleagues started with cost data gathered in Zimbabwe and then modelled this for each country using variables related to, for instance, local salaries and supply costs. From this they estimated that the total of direct and indirect costs to provide one circumcision, a unit cost, would be about $95 in South Africa and $65 in Uganda (Njeuhmeli et al., 2011).

In calculating the impact, Njeuhmeli and the team needed to gather and use country specific data related to each of the following: demography, sexual behaviour and HIV prevalence trends. They also needed to review and revise economic, epidemiological and MC effectiveness assumptions. The DMPPT presents a default value of $7,400 as the discounted lifetime cost of ART. This is the value that can be saved from one HIV infection averted (Bollinger, DeCormier Plosky and Stover, 2009). When these data are entered and modelled, the results generated indicate the number of MCs required for each HIA, the net cost per HIA and the cost savings per HIA.

DMPPT Modelling for Uganda

This 2011 modelling for Uganda showed that in order to reach 80% MC prevalence by 2015 it needed to perform 4.25m MCs and an additional 2.1m in the years 2016 to 2025 to maintain that degree of prevalence. The total discounted cost of doing so was estimated at just over $350m. The number of MCs required to gain one HIA varies from country to country; this modelling indicated that Uganda needs 19. This suggests that such a programme would avert 350,000 new infections, 25% of the total expected over that 14 year period, at a cost of $1,408 for each HIA, which in turn would generate future savings of $5,992. In summary, 6.35m MCs at a discounted cost of $350m, would generate total future savings of $1.52 billion (Njeuhmeli et al., 2011).

Whilst MC impacts first on the risk of infection for the circumcised man, the DMPPT shows that in a scaled-up programme, as modelled by Njeuhmeli et al., by 2025 nearly half of the infections averted are in women. All of the research and modelling to this point in time, November 2011, shows without any doubt that male circumcision is a very cost-effective intervention that promises substantial impact on population-wide HIV infection rates. In the Editor’s Summary, countries and donors are encouraged to adopt MC and to rapidly implement scaled-up programmes, ensuring the careful allocation and deployment of sufficient resources. The potential impact that MC can have on slowing the pace of the AIDS epidemic, and the future savings that can be made, are just too great to be missed (Njeuhmeli et al., 2011).

It had taken 4 years from the 2007 WHO MC recommendation for Njeuhmeli et al. to develop the tools and generate the necessary data to determine with some certainty the costs involved, and the potential impact that could be made, if a country deployed this, still, innovative prevention intervention. One might have been expecting, or at least hoping, that each of these 13 countries would do exactly that, embrace MC as an essential prevention intervention and deploy key resources to ensure the rapid move towards fully scaled-up programmes. Now in the fourth quarter of 2012, each country should be well on its way to achieving the initial 2015 catch-up targets. This however does not seem to be the case and it is important to learn why.

Performance Against DMPPT Targets by March 2012

Data released by PEPFAR in March 2012, shows that in the 5 years from the WHO and UNAIDS MC recommendation, in these 13 countries, a total of 1.2m boys and men had been circumcised by organisations supported by PEPFAR funding. As PEPFAR is by far the largest contributor to HIV funding in the region, including budget allocations from country governments, this probably represents 80% – 90% of the total numbers performed. Further research is needed to show more clearly the total numbers being performed regardless of source of funding. It would be helpful for programme planners, government ministries and donors to know, with certainty, the true scale of the numbers being reached. Extracting data for just those circumcisions provided to the target population of men aged 15 years and above, and only procedures performed in the 13 countries in the period October 2009 to March 2012, a total of 901,900 circumcisions are reported, 4.3% of the 2015 target, 20.8m. Performance against target varies widely from country to country; Kenya has performed the largest overall number, 295,800, which is also the greatest percentage of a country target, 34.4%. Uganda has performed the second largest number, 149,400 but this represents just 3.5% of its 4.2m target (Reed et al., 2012).

In reviewing what may have contributed to Kenya’s successful performance, the authors draw attention to two key factors. The government of Kenya moved very quickly after the WHO recommendations in 2007. It immediately started active engagement with community elders and worked to ensure their full and active support for the MC programme and in mobilising their communities. This led to creating the new term, Voluntary Medical Male Circumcision (VMMC), in which the words voluntary and medical were deemed to be important for all and made it easier to gain consensus support for the programme. Like most other countries in this region, Kenya does not have sufficient numbers of doctors or clinical officers to satisfy the required programme volumes. So in an innovative move, Kenya passed changes to its national task-shifting policy to allow specifically trained non-physician clinicians to perform VMMC (Reed et al., 2012).

These countries have been faced with common global and regional issues, including economic downturn, national elections, military conflicts and the cancellation of Global Fund round 11, all of which may have caused progress to be slower than hoped for. All can see the significant future savings that a scaled-up MC programme might deliver, but these savings are in the future, after the current term of office for those in power. It may be difficult to give sufficient time and focus to a health intervention such as MC, the benefits of which may not become apparent for a long time, compared to more immediate and tangible issues being faced (Reed et al., 2012).

Target of 4.7m MCs by end of 2014 Supported by Obama

On World AIDS Day 2011, December 1st, President Obama pledged support to enable this region to have performed a total of 4.7m circumcisions by the end of 2014 (Reed et al., 2012). Whilst that is quite small compared to the previously set target of 20.8m by the end of 2015, it would represent scaling-up of MC annual activity across the region by a factor of three times, compared to performance in 2011-12. The Uganda AIDS Commission (UAC) has set a target of 1.25m circumcisions by the end of 2013 (Uganda AIDS Commission, 2011). Doing so will require Ugandan service providers to perform 1m circumcisions in the next year; that would need providers to scale-up by a factor of about 4 times

What must be done in order to achieve these targets, as set by Obama and UAC? The learning so far would indicate a need for more focus and attention on engaging political and community leaders, possible changes to national task-shifting policies, adoption of multiple and varied operational models including fixed, mobile, outreach and deploying some staff on a full-time dedicated basis and others part-time in which MC is blended into the business as usual.

WHO Announces an Innovative Non-Surgical Circumcision Device

Whilst there is a need to consider ways to improve political and community engagement, develop more efficient operational models and address the issues presented by the lack of available, skilled resources, some progress is being made in applying technological innovation to the circumcision procedure. In March 2012 the WHO issued a report of the findings from the January meeting of the Technical Advisory Group on Innovations in Male Circumcision (TAG). This noted that an elastic ring controlled radial compression device, PrePex, had completed trials in Rwanda and was now approved for use, in Rwanda, as an alternative alongside the established surgical methods for male circumcision. It noted that PrePex can be used by physicians and non-physician clinicians. PrePex has two key parts, a plastic ring which is inserted inside the foreskin and a rubber ring which is placed on the outer foreskin such that it compresses the foreskin against the inner plastic ring, stopping the blood flow. The foreskin below where the rings have been placed becomes necrotic and can be removed after 7 days. The procedure requires no local anaesthesia and no suturing is necessary after the cutting and removal of the necrotic skin. The TAG report suggests that such innovations are useful if these enable services to be provided more quickly, with improved safety, possibly at lower cost and make it easier for implementers to use lower cadres of staff who are often more available in larger numbers. If some or all of these benefits apply and the innovation is acceptable to both the client and the provider, then it may contribute to a greater, easier acceleration of the programme towards achieving necessary targets (World Health Organisation, 2012).

The TAG reached its findings after completion of three trials in Rwanda, these being necessary to meet the requirements of the WHO defined pathway for clinical evaluation of such medical devices. The first assessed the safety and efficacy of PrePex in performing circumcision (Bitega et al., 2011). A second compared the use of PrePex to a surgical method for circumcision (Binagwaho et al., 2011). The third assessed the relative safety and AE rate when the procedure was performed by non-physician clinicians compared to physicians (Mutabazi et al., 2012).

These clinical trials proved that PrePex is efficacious in performing male circumcision and that it can be safely used by non-physician clinicians. The trials did not provide any usable operational data regarding either the unit cost of performing a circumcision with PrePex or how it might improve the efficiency of an operator, though there were some indications that the procedure takes less time than the surgical method. This research is needed to provide data to fill in these gaps.

In order for PrePex to be fully approved by WHO for use in other countries, results from 3 similar trials in a second country are necessary. These trials are almost complete in Zimbabwe and the TAG is expected to report again in January 2013. Whilst waiting for this approval, the WHO states clearly that other countries should not conduct similar comparative studies but instead conduct preliminary studies into the local acceptance and feasibility of deploying PrePex. The TAG notes the need for further studies to determine the most efficient and safe operational models which might enable non-physicians to be engaged in the deployment of PrePex (World Health Organisation, 2012).

Recent research, in press and yet to be published, reviews the early lessons from a Safe Male Circumcision programme in Uganda. This indicates that, in a private hospital setting, with specific training and close supervision, non-physician clinicians are able to safely perform 15 circumcisions per day using the sleeve resection method (Galukande et al., 2012).


AUVERT, B. et al., 2005. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 Trial. PLoS medicine, 2(11), p.e298.

BAILEY, R.C. et al., 2007. Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet, 369(9562), pp.643–56.

BITEGA, J.P. et al., 2011. Safety and efficacy of the PrePex device for rapid scale-up of male circumcision for HIV prevention in resource-limited settings. Journal of Acquired Immune Deficiency Syndromes, 58(5), pp.e127–34.

BINAGWAHO, A. et al., 2011. A prospective, randomized, open-label trial comparing the PrePex™ device to surgical circumcision for rapid scale-up of male circumcision in resource-limited settings (Protocol Number RMC-01). Kigali, Rwanda, 2011.

BOLLINGER, L., DeCORMIER PLOSKY, W. and STOVER, J., 2009. Male Circumcision: Decision Makers’ Program Planning Tool, Calculating the Costs and Impacts of a Male Circumcision Program. Washington, DC: Futures Group, Health Policy Initiative, Task Order 1.

CAMERON, D.W. et al., 1989. Female to male transmission of human immunodeficiency virus type 1: risk factors for seroconversion in men. Lancet, 2(8660), pp.403–7.

FINK, A.J., 1986. A possible explanation for heterosexual male infection with AIDS. The New England journal of medicine, 315(18), p.1167.

GALUKANDE, M., BBAALE, D., COUTINHO, A., et al., 2012. Mass Safe Male Circumcisions: early lessons from a Ugandan urban site – A case study. (submitted for publication)

GRAY, R.H. et al., 2007. Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet, 369(9562), pp.657–66.

HALPERIN, D.T. & BAILEY, R.C., 1999. Male circumcision and HIV infection: 10 years and counting. Lancet, 354(9192), pp.1813–5.

HILL, A.B., 1965. The Environment And Disease: Association Or Causation? Proceedings of the Royal Society of Medicine, 58, pp.295–300.

KAHN, J.G., MARSEILLE, E. & AUVERT, B., 2006. Cost-Effectiveness of Male Circumcision for HIV Prevention in a South African Setting. PLoS Medicine, 3(12), p.10.

KUZNIK, A. et al., 2012. Medical male circumcision for HIV/AIDS prevention in Uganda – the cost of disposable versus re-usable circumcision kits. Tropical Doctor, 42(1), pp.5–7.

MOSES, S. et al., 1990. Geographical Patterns of Male Circumcision Practices in Africa: Association with HIV Seroprevalence. International Journal of Epidemiology, 19(3), pp.693–697.

MOSES, S., PLUMMER, F.A., et al., 1994. The Association Between Lack of MC and Risk for HIV Infection. A Review of the Epidemiological Data. Sexually Transmitted Diseases, (July-August), pp.201–210.

MURRU, M. et al., 2003. Costing health services in Lacor Hospital. Health Policy and Development, 1(1), pp.61–68.

MUTABAZI, V. et al., 2011. One arm, open label, prospective, cohort field study to assess the safety and efficacy of the PrePex™ device for scale-up of non-surgical circumcision when performed by nurses in resource-limited settings for HIV prevention (Protocol RMC-03). Kigali, Rwanda, 2012.

NJEUHMELI, E. et al., 2011. Voluntary medical male circumcision: modeling the impact and cost of expanding male circumcision for HIV prevention in eastern and southern Africa. PLoS Medicine, 8(11), p.e1001132.

ODAGA, J. and OKIRIA, C., 2010. Providing CD4 Cell Count Tests to Hard-to- reach Communities in Northern Uganda Programme Reach and Cost-effectiveness of an Outreach Delivery Model.

REED, J.B. et al., 2012. Voluntary Medical Male Circumcision: An HIV Prevention Priority for PEPFAR. JAIDS, 60(August), pp.88–95.

SHEPARD, D.S., HODGKIN, D. & ANTHONY, Y., 1998. Analysis Of Hospital Costs: A Manual For Managers.

UGANDA AIDS COMMISSION, 2011. The National Priority Action Plan.

UNAIDS, 2012. Together we will end AIDS. Available at:,72588,en.asp [Accessed August 1st, 2012].

UNAIDS, WHO and SACEMA, 2009. Male circumcision for HIV prevention in high HIV prevalence settings: what can mathematical modelling contribute to informed decision making? PLoS Medicine, 6(9), p.e1000109.

WEISS, H.A., QUIGLEY, M.A. & HAYES, R.J., 2000. Male circumcision and risk of HIV infection in sub-Saharan Africa: a systematic review and meta-analysis. AIDS, 14(15), pp.2361–2370.

WILLIAMS, B.G. et al., 2006. The Potential Impact of Male Circumcision on HIV in Sub-Saharan Africa. PLoS Medicine, 3(7), p.e262.

WORLD HEALTH ORGANISATION, 2010. Considerations for implementing models for optimizing the volume and efficiency of male circumcision services. [Kim Eva Dickson (WHO), Jason Reed (CDC, USA) and Dino Rech (PSI/WHO Consultant)].

WORLD HEALTH ORGANISATION, 2011. Joint Strategic Action Framework to Accelerate the Scale-Up of Voluntary Medical Male Circumcision for HIV Prevention in Eastern and Southern Africa.

WORLD HEALTH ORGANISATION, 2012. Use of devices for adult male circumcision in public health HIV prevention programmes.

WORLD HEALTH ORGANISATION, 2012a. World Health Statistics 2012. Available at: [Accessed September 26th, 2012.]

WORLD HEALTH ORGANISATION and UNAIDS, 2007. New Data on Male Circumcision and HIV Prevention: Research Implications for Policy and Programming. Montreux, Switzerland.