Circumcision could have beneficial effect regardless of change in condom use

Gus Cairns
Published: 16 August 2006

If more than 63% of potential HIV infections are stopped by the widespread adoption of male circumcision – a rate consistent with that seen in the first randomised controlled study presented last year – then there would be a net positive effect on HIV incidence in the population even if a 50% rate of consistent condom use went down to zero. This, at least, would be the situation in Soweto, South Africa, as predicted by mathematical modelling done by Kyeen Mesesan of the Yale University department of Public Health in data presented to the Sixteenth International AIDS Conference in Toronto.

Mesesan based her model on the results seen in the study by Bertrand Auvert, whose participants were men from Orange Farm, a township near Soweto. She said that if the protective effect of circumcision was ignored there would be 318,000 new HIV infections among the 823,000 sexually-active adults in Soweto over the next 20 years. The overall HIV prevalence would increase from 16 to 23%.

If however you assumed a 35% rate of circumcision amongst the male population – that seen at Orange Farm - then 20-year prevalence goes down to 17% - a 1% increase – and the number of new infections over that period decreases 23% to 244,000. If an additional 20% of men were circumcised in a mass prevention programme, then an additional 53,000 infections would be prevented – a further 22% decrease – and the 20-year prevalence rate would be 14%, or 2% lower than it is today.

If overall condom use (currently about 50% in South Africa, according to other surveys) decreased to 35%, then such a mass programme would only prevent 18,000 new infections. However the protective effect of circumcision is such that if the protective effect seen in the intent-to-treat analysis of the Orange Farm (a 61% reduction in infections) is replicated in the other randomised controlled studies still continuing, then condom use could fall to 3% before the benefit of such a 20% circumcision programme was completely lost. If the protective effect was 63% all men could stop using condoms and there would still be a neutral effect on new HIV infections.

Jim Kahn of the University of California, San Francisco said that assuming a cost per circumcision of $55.75, the cost of a mass male circumcision programme where HIV prevalence was 8.4% - regardless of the programme’s actual coverage - would be $550 per HIV infection averted, but that when antiretroviral and other treatment costs were taken into account there the efficacy of circumcision would be a net saving of $753 per circumcision. He said that circumcision would be a cost-saving measure at any efficacy above 21%.

Mesesan's and Kahn’s models were criticised by audience members as being if anything too pessimistic; the ‘on-treatment’ efficacy observed in the Orange Farm study was above 70%, and Kahn’s scenario fails to take into account the ‘herd immunity’ effect, of there being an accumulating effect of fewer HIV-positive men in the population as time went on.

These projections assume that the protective effect seen in the Orange farm study is replicated in other studies. Robert C Bailey of the Chicago University School of Public Health presented interim data from the trial currently underway in Kisumu, Kenya, among young men aged 19-24. He said that 2,784 men had been randomised out of an estimated young male population of 39,000 in the city. The young men were mainly from the Luo tribe, a group that has low levels of circumcision (8% in Kisumu) and high HIV prevalence compared with other peoples in the area.

Consistent condom use was quite low at only 28% and only 6% of the men were married or had a live-in partner. Four per cent said they’d had anal sex and only four men admitted to having sex with men. This was clearly a crucial age group to conduct the trial in as HIV prevalence in 18 year-old men was only 1% but by age 24 had risen to 17%, indicating an annual incidence of 3.5%. HIV prevalence among men screened for the trial had been 8% overall.

Recruitment is almost completed for the trial and results will be complete in 2008. So far 1,334 out of 1,391 randomised to circumcision have had the operation, with an adverse event rate (largely delayed healing and swelling) of 1.7%. Condom use by participants had increased after circumcision. The proportion reporting consistent condom use increased in the three months after the operation from a baseline of 22% then stayed the same, and was 35% 24 months after circumcision. The proportion who reported condom use at their last sex similarly increased from 48% at baseline to 60% at 24 months.

One worrying factor was that the annual HIV incidence so far observed in trial participants was 1.8%. This was lower than the forecast incidence of 2.5% and meant, according to Bailey, that the study would ‘just about’ have enough statistical power t prove efficacy. Another worry was that despite instructions not to have sex for 30 days after circumcision, 10% of those circumcised actually did so, though most started near to the 30-day limit.

Bailey said that 99% of female partners were ‘very satisfied’ with their partner’s circumcision, then raised some laughter when he revealed that this was in fact what the young men said about their girlfriends’ reactions.

Circumcision programmes will vary in effectiveness according to the proportion of the male population already circumcised, and may have to be carefully targeted. A survey of tea plantation workers in the Southern Rift Valley province several hundred kilometres east of Kisumu found that over 80% of male workers were in fact circumcised, and the vast majority of those who weren’t were members of the Luo tribe. Three-quarters of them had had traditional circumcision rather than in healthcare settings.

Lastly, a prospective study looked as possible behaviour change in 648 Luo men in western Kenya who presented themselves at healthcare settings wanting circumcision. They were randomised into 324 who had immediate circumcision and 324 who agreed to wait a year.

Kawango Agot of the Impact Research and Development Organisation, a local NGO conducting health research, drew applause when she said of circumcision: “If it’s not helpful to women it’s not helpful at all.” She surveyed the reasons men came forward for circumcision and found that the most common reason was an already-existing perception among men and especially among their partners that circumcision was more hygienic. Men who had got circumcised had had a higher number of sexually transmitted infections in the past. Another reason was sexual dysfunction: a high proportion of men coming forward for circumcision had erection problems of one sort or another, and circumcision was seen as a last-resort attempt to correct these. The survey found that after the first month there was no difference in the proportion of men who had had sex, casual sex, or sex without a condom, between the circumcised or uncircumcised groups.

Despite its claimed benefits in reducing HIV incidence, the introduction of any mass circumcision programme is clearly not going to be without its critics. Several members of the audience continued to regard circumcision at ‘mutilation’ or saw attempts to introduce it as an essentially coercive method of reducing HIV incidence akin to quarantine (despite reports that in most surveys a majority of African men – up to 80% - said they would be willing to be circumcised if it was protective against HIV). Other audience members were concerned that announcements about the effectiveness of circumcision (and, by extension, other new prevention technologies) would confuse people and cause them to abandon condom use.

The one charge that proponents of the new prevention technologies are perhaps vulnerable to is that in most cases (with one exception on the circumcision panel) those conducting the programmes are white public health experts and researchers from foreign NGOs and institutes, underlining the importance of continued community consultation in all these trials of new prevention approaches.


Mesesan K et al. The potential benefits of expanded male circumcision programs in Africa: predicting the population-level impact on

heterosexual HIV transmission in Soweto. Sixteenth International AIDS Conference, Toronto, abstract TUAC0203, 2006.

Kahn JG. Cost-effectiveness of male circumcision in sub-Saharan Africa. Sixteenth International AIDS Conference, Toronto. Abstract TUAC0204. 2006.

Bailey RC et al. A randomized controlled trial of male circumcision to reduce HIV incidence in Kisumu, Kenya: progress to date. Sixteenth International AIDS Conference, Toronto. Abstract TUAC0201. 2006.

Sateren WB. Male circumcision and HIV infection risk among tea plantation residents in Kericho, Kenya: incidence results after 1.5 years of follow-up. Sixteenth International AIDS Conference, Toronto, abstract TUAC0202, 2006.

Agot K. Male circumcision in Siaya and Bondo districts, Kenya: a prospective cohort study to assess behavioural disinhibition following circumcision. Sixteenth International AIDS Conference, Toronto, abstract TUAC0205, 2006.

Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.