Male Circumcision and HIV

Circumcision offers no protection against STIs; UNAIDS blogs two studies from last year

The Journal of Pediatrics is reporting in their March 2008 edition that "early childhood circumcision does not markedly reduce the risk of the common [sexually transmitted infections] in the general population in [developed] countries."

They are hedging a bit, as their conclusion sounds as if some reduction was observed. Just not enough. It may be the stilted nature of scientific language, but they are clearer in the abstract that "STIs were not statistically significantly different-23.4 and 24.4 per 1000 person-years for the uncircumcised and circumcised men, respectively." [Emphasis mine.]

Reference

Dickson NP, van Roode T, Herbison P, Paul C., J Pediatr. 2008 Mar;152(3):383-7. Epub  2007 Oct 22

The UNAIDS blog, HIV This Week, decided to repeat, perhaps for filler? to beat back lagging interest? a couple of studies that came out late last year.

The first one found after a lot of blather about more studies being needed, "In conclusion, after adjustment, male circumcision was not significantly associated with women’s HIV risk." [Emphasis theirs.]

Edit: Does this mean that the prevalence of HIV was equal as between the circumcised and intact groups? Or could it be that the act of heterosexual intercourse neutralizes the protective effect of circumcision? Or rather does this study call into question all the circumcision RCTs? Bet on the first and third possibilities.

Reference

Turner AN, Morrison CS, Padian NS, Kaufman JS, Salata RA, Chipato T, Mmiro FA, Mugerwa RD, Behets FM, Miller WC. Men’s circumcision status and women’s risk of HIV acquisition in Zimbabwe and Uganda. AIDS 2007;21:1779-89.

The second study was a systematic review of the literature that looked at other issues but recorded circumcision status. The authors talk shit and then admit the obvious. After noting a suggestion of "a strong association between male circumcision and lower HIV among men who have sex with men," they conclude,
"The observed effect might be due to confounding factors not measured (and therefore not controlled for) in the studies, rather than being the result of a biological effect of male circumcision."

Reference

Fankem SL, Wiysonge CS, Hankins CA. Male circumcision and the risk of HIV infection in men who have sex with men. Int J Epidemiol 2007 Oct 19; Epub ahead of print.

I couldn't find a way to easily link to this blog entry, so go here and good luck.

Reuters: Circumcision does not affect HIV in U.S. men

Reuters is reporting that circumcision does not protect against HIV in black and Latino men, consistent with the Sydney study, in the first media-wide reports on this new survey.

Unfortunately, the reporter can't resist repeating unproven speculation as to why an intact penis has been cited to be dangerous, citing unlikely methods of entry through "tears" during intercourse.

Link: Circumcision does not affect HIV in U.S. men: study | Health | Reuters.

"Risk reduction" strategies no substitute for consistent condom use, study shows

Ok. Admittedly, condoms are themselves a form of risk reduction. But in my view, at a high 90% effectiveness rate, condoms function more like a vaccine. Ok. That's a bad comparison, too. But the point is that so-called risk reduction strategies, such as sero-sorting, only engaging in the insertive role, and choosing partners according to undetectable viral load, are hardly any protection at all in comparison.

So what of circumcision?

It seems more likely than not that circumcision falls in there with strategies most similar to choosing to engage in only the insertive role. A rather pure form of gambling.

While the message has been that risk reduction strategies (including male circumcision) may have some importance across populations, the researchers conclude that "risk reduction strategies seem to fail to prevent HIV infection on an individual level."

Why is this an important point? Because this is precisely the opposite message delivered by Bailey, Halperin and others concerning implementation of circumcision in the United States and other developed countries. How often have they said to the media that mass circumcision may not have value for low prevalence countries, but individually it could save your life?

It is becoming clearer that it has no value in developed countries as a risk reduction strategy either for whole societies or individuals.

Edit: And I might add this study reinforces an emerging picture supported by the Sydney study that showed circumcision was of no value in halting the spread of HIV among gay men.

Aidsmap article after the jump.

More evidence that condom use a primary source of stable HIV incidence

HIV negative men who engage in high risk sexual behavior but who also engage in sero-sorting, or the practice of choosing partners of the same HIV status, sero-convert at much greater rates than men who use condoms as their primary method of prevention, so reports a new study from The Netherlands.

When I read this article, it occurred to me that both sero-sorting and condom use would provide much greater protection than either alone. But then I realized that if you are using condoms, at an effectiveness rate of around 99%, sero-sorting would yield a negligible gain. So ultimately, condoms are the answer and everything else is a distant second.

An echo of this phenomenon can easily be imagined in the case of circumcision. HIV negative men who depend on circumcision as a method of prevention will sero-convert at much greater rates than condom users. Obviously. Give them condoms and circumcision simply has no effect at all on sero-conversion rates. Wasn't this the problem with the circumcision/HIV studies?

At any rate, the point here is, given condoms or discussing and choosing partners based on HIV status or in many cases perceived status, condoms win the effectiveness race every time. Given condoms or circumcision, the choice is equally obvious. Circumcision posits a hopelessly confused and confusing message that will not work and sets the stage for "prevention" based on wishful thinking and denial, and ultimately a worse epidemic.

Aidsmap fair use article after the jump.

Australian AIDS org says circumcision "has no role" in Australian context

The Australian Federation of AIDS Organisations has issued a briefing paper [html] with the telling headline, "Male circumcision has no role in the Australian AIDS epidemic."

Among the key points are:

• There is no demonstrated benefit of circumcision in men who have sex with men.
• Correct and consistent condom use, not circumcision, is the most effective means of reducing female-to-male transmission, and vice-versa.
• African data on circumcision is context-specific and cannot be extrapolated to the Australian epidemic in any way.

The first point is a reiteration of a recent study from Sydney that showed circumcision provided no protection for gay men. The second point is clear from the statements and caveats issued from the circumcision studies organizers themselves. The third point is clear from the key contextual attribute of all three studies, that the protective effect is in heterosexual couples in only one direction.

A high impact, succinctly written poster presentation given at this years's International AIDS Society Conference is available here [large pdf] and is worth printing in color for outreach activities.

Download Circumcision07.pdf

Women with circumcised partners at greater risk for STIs and derive no protection viz HIV, study shows

The first African study on the effect of male circumcision on women's risk for STIs and HIV has trickled out. And the news is not encouraging for those hoping circumcision would provide a magic bullet equally between the sexes for women. 

Aidsmap reports, "Consistent with the baseline findings, women partnered with circumcised partners had higher levels of sexual risk during follow-up, being more likely than women with uncircumcised partners to self-report a sexually transmitted infection (6% versus 4%, p < 0.001), have symptoms of a sexually transmitted infection (26% versus 20%, p < 0.001), and to have a risky sexual partner (a man with symptoms of a sexually transmitted infection, or who was HIV-positive; 23% versus 14%, p < 0.001). Women with circumcised partners also had a lower mean number of protected sex acts (8.6 versus 8.3 per month, p < 0.001)."

A small,statistically insignificant protective effect was observed for HIV, "But the protective effect of having a circumcised partner weakened when the investigators adjusted their model to include age, age at sexual debut, contraceptive use, husband’s employment status, level of education, and number of sexual partners in the previous three months (HR, 0.78; 95% CI, 0.53 – 1.14, p value not provided). Further adjustment, taking into account population subgroups, meant that the protective effect of circumcision disappeared entirely (HR, 1.03, 95% CI, 0.69 – 1.53) [emphasis mine]."

The ultimate  conclusion as between circumcised and intact partners of women was that, "The investigators found that low-risk Ugandan and Zimbabwean women had a similar risk of HIV infection, regardless of their partner’s circumcision status. However, the high-risk Ugandan women derived a non-significant protective effect from having a circumcised partner (HR 0.16; 95% CI, 0.02 – 1.25) [emphasis mine]."

This study is consistent with a number of other studies that have shown little difference in HIV infection rates between groups of similar cultural and sexual mores, usually meaning comparisons within countries as opposed to cross-border comparisons.

Full Aidsmap summary after the fold.

Of receptor sex, foreskin interventions, and disappointed scientists

A reader writes in regarding an article on Medscape worth noting. It isn't that the findings are so earth-shattering. Rather it's that the article's author and apparently the researchers and commentators as well appear to want so badly for circumcision have some value. But apparently for gay men it offers nothing -- even in a study where the researchers clearly were hoping it would.

You can almost feel one Matthew Hogben's sullen mood, glimpsing his leaden face, eyes cast downward in disappointment, when he's says, "The Australian study showed that circumcision was not a factor one way or the other in whether [individuals] developed HIV, which probably makes sense."

The logical thing to do when a study fails to show any beneficial effect would be to abandon the avenue and look elsewhere. But not these scientist-advocates. Hogben says, "with the findings of the trials that ran in Africa, [researchers] might want to expand study of the association to see to what extent they could take advantage of those findings to look at secondary effects. It might be worthwhile to see whether there's less transmission or 'burden' with a foreskin intervention, but I suspect we would need more data, and more clean data, to confirm this."

Translation: This study failed. Let's do more.

These days, foreskin interventions are all the fashion.

THANK YOU to Scot Anderson for this contribution.

A fair use glimpse of Medscape article after the jump.

Sociodemographic and behavior/risk characteristics explain "effectiveness" of circumcision

In a long description of a recent study investigating the (real world) effectiveness of circumcision, Aidsmap has highlighted important data that reveals an association with sociodemographic (aka cultural and value-based) characteristics and HIV infection risk. The money quotes are as follows.

... Circumcision was significantly associated with tribal affiliation [hence, the common practice among a single group], high school education [perhaps indicating greater knowledge of the perils of unprotected sex], fewer marriages [a likely lower incidence of concurrent partnerships or greater emphasis on serial monogamy?], and a smaller age difference between spouses [indicating a greater power balance within marriage].

...

Since the majority of uncircumcised men belonged to the Luo tribe, sociodemographic and behavioural/HIV risk characteristics and HIV incidence rates between Luo and non-Luo males were investigated. Luo males were significantly older [indicating a longer sexual history], more likely to practice traditional African religions, were significantly older than their spouses by more than 10 years, and reported sex with a commercial sex worker. Regardless of the circumcision status, Luo men were 4.6 times more likely to become HIV-infected.

The largely American researchers, possibly exhibiting their pro-circumcision bias, came to the perplexing conclusion, "The findings provide additional evidence that circumcision by traditional circumcisers offers protection from HIV infection in adult men in rural Kenya where circumcision is common."

A more logical conclusion would be that differences in the ages of spouses, a history of using the services of commercial sex workers, generally more permissive behavioral patterns in sexual relations, and greater power differences between spouses are indicative of greater HIV infection risk. And this we already knew.

Fair use Aidsmap story below the fold.

Circumcision is not predictive of HIV infection (seroconversion) in gay men.

But not content with these results, the authors conclude, "More studies, please."

Edit: Aidsmap has an excellent summary of the study and one other study showing Peruvian gay men would be willing to be test subjects for a future circumcision study in that country. Obviously, the first study would seem to preclude any need for making guinea pigs out of South American men.

Abstract after the break.

Sound Advice A Measured Response

I found this blog post from the Hong Kong based Health Reporter, a publication of the Journalism and Media Studies Centre at the University of Hong Kong that I thought worth acknowledging. Click through to read it on their site.