Gotta love the circumcision lobby's tenacity and hate the AIDS publishing industry's gullibility on this one ...
That's right. Under a headline that concludes circumcision protects "some" gay men, the text provides us with these conclusive quotes:
The study also showed that overall circumcision did not protect gay men from infection with HIV. Earlier data from the study presented to the conference of the International AIDS Society in Sydney in 2007 showed that circumcision had no protective effect.
...
Studies in Africa have shown that circumcision reduces the risk of HIV infection for heterosexual men. However, a recent meta-analysis found no conclusive evidence that circumcision was protective for gay men.
...
Statistical analysis that included the entire study population showed that circumcision did not provide any significant protection against infection with HIV.
Only 10% of the study’s person years of follow-up was contributed by men who reported insertive unprotected sex but not receptive sex without a condom. There were only four HIV infections in these men. Analysis showed that circumcised men who only reported insertive unprotected sex did not have a significantly reduced risk of HIV.
...
So how did these half-wit scientists conclude that circumcision was associated with lower incidence of HIV infection in these gay men?
Next the investigators restricted their analysis to men who stated a preference for the insertive position in all anal intercourse. These 435 men (279 of whom were circumcised) contributed 1710 person years of follow-up.
There were a total of seven HIV infections in these men, five of which were in the uncircumcised men.
Statistical analysis showed that circumcision was associated with a significant reduction in the risk of HIV infection for men with a preference for insertive anal sex (p = 0.049). This association was strengthened when the investigators adjusted for age and potentially serodiscordant unprotected anal intercourse.
They admit in the end that the evidence is weak and designing studies to provide strong evidence would be nearly impossible to design for many reasons, but characteristically omit any mention of ethical barriers.
Reference
Templeton DJ et al. Circumcision and risk of HIV infection in Australian homosexual men. AIDS 23: 2347-51, 2009.
Link: Circumcision protects gay men who have a 'preference' for insertive sex from HIV
