A panel of experts organized by UNAIDS, WHO and the South African Centre for Epidemiological Analysis have calculated the numbers needed to treat (NNT) for any roll out of mass circumcision and published the result in the open access journal PLoS Medicine.
The wildly optimistic projection is that as few as five circumcisions will be necessary to prevent one already preventable case of HIV infection in men in high prevalence countries.
The group also concluded that even if circumcised men either reduced their use of condoms or resumed sex too soon after the operation, circumcision would remain beneficial on a population level. They also concluded that women will indirectly benefit from circumcision.
There is ample evidence to the contrary to the forgoing here and here. Moreover, according to Aidsmap, "The published paper does not contain detailed numerical projections of the impact of circumcision in various circumstances, and concentrates on the situation in the highest prevalence countries." Hence, their conclusions are unlikely to apply, to the extent they are accurate, to countries anywhere else in the world outside of "Zimbabwe, Zambia, Botswana, Namibia and South Africa."
As recently as July of this year, the cost effectiveness of circumcision was calculated to be a distant third in a much more thorough study conducted by the British Columbia Centre for Excellence in HIV/AIDS.
Link: UNAIDS estimate that one infection will be avoided for every 5 to 15 men circumcised
Citation: UNAIDS/WHO/SACEMA Expert Group on Modelling the Impact and Cost of Male Circumcision for HIV Prevention (2009) Male Circumcision for HIV Prevention in High HIV Prevalence Settings: What Can Mathematical Modelling Contribute to Informed Decision Making? PLoS Med 6(9): e1000109. doi:10.1371/journal.pmed.1000109
EDITS: I have made some minor edits without adding or changing the forgoing content.
