Where are those African studies?

[by-the-lake]

I have a doc telling all kinds of students about 2 different independent studies in Africa that circs reduces HIV. I have heard of these, and would like more info to present to him, because I heard they were not right. Info?

[AnnDMFT]

Here is a link that goes through and discusses all of the flaws in these studies....

I am interested in seeing if anyone can get some from other places too!

http://www.cirp.org/library/disease/HIV/

[Ron_Low]

There are THREE large Randomized Controlled Trials from Africa in the last decade which collectively imply that voluntary adult circumcision may correlate with a roughly 60% relative reduction in female-to-male HIV transmission. 

The first was South Africa.  It was the smallest and poorest controlled, and showed the strongest effect. 

Then there was Kisumu Kenya (Bailey) and Uganda (Wawer/Gray). 

All three trials were stopped early (after about 25% of the pre-agreed duration), which magnified the fact that HIV can't be measured in the blood sometimes until months after infection.  This might not have been a confounder except that the cut men were abstinent for about 6 weeks after surgery during which no f-to-m transmission could occur.  It was rolling enrollment so at the early conclusion some men had been involved for only a few months.

ALL the studies shared the design drawbacks discussed at Circumstitions.  To me the biggest flaws are:

- NOT random.  Every subject sought circumcision for cultural reasons and so was perhaps pre-disposed to find value to it or change to more mature behavior or status after the cut.  Subjects were also paid, so there was a potential desire to please benefactors/researchers involved. 

- NOT blinded at all.  Cut men had multiple follow-up visits at the clinic with safe-sex counseling and arm-loads of condoms each time. 

- MANY more subjects lost to follow up than were infected, in either treatment or control groups.  How can we assume these men were infected at the same rate as the remaining subjects?  They might have all gotten sick and sought help at a real hospital.  Or perhaps their circumcisions absessed and they sought treatment elsewhere. 

- NO concordance with the real world.  Half of African nations have higher HIV incidence among the circumcised. 

- REFUTED by the researchers themselves: In 2009 Wawer / Gray reported that the Ugandan men they circumcised were 50% MORE likely to infect their female partners with deadly HIV.  The M-to-F effect washes out any F-to-M benefit, but the WHO has not called for more study to confirm this finding (which is good since it was an immoral Tuskegee-esque atrocity).  In 2010 Bailey followed up and confirmed that men in Kisumu, Kenya were no less likely to be HIV positive if circumcised. 

- IRRELEVANT to infants, who don't have sex. 

- RENOUNCED as human rights abuses by recently updated national medical association policies in Australia and Holland.  They say even if true the studies don't justify infant circumcision and don't justify adult circumcision outside of the regions studied. 

[Ron_Low]

See also:

"Not All RCTs are Created Equal"

20 January 2010
 

by

Robert S. Van Howe,
Marquette, MI, USA
Clinical Professor of Pediatrics, Michigan State University College of Human Medicine
- - - - -

I believe you'll have to Google for it, as MDC policy doesn't favor copying others' content. 

{here is a link: http://www.annfammed.org/cgi/eletters/8/1/64#11574 }

-Ron

[azgirl]

Thank you!
 

Originally Posted by Ron_Low 

There are THREE large Randomized Controlled Trials from Africa in the last decade which collectively imply that voluntary adult circumcision may correlate with a roughly 60% relative reduction in female-to-male HIV transmission. 

The first was South Africa.  It was the smallest and poorest controlled, and showed the strongest effect. 

Then there was Kisumu Kenya (Bailey) and Uganda (Wawer/Gray). 

All three trials were stopped early (after about 25% of the pre-agreed duration), which magnified the fact that HIV can't be measured in the blood sometimes until months after infection.  This might not have been a confounder except that the cut men were abstinent for about 6 weeks after surgery during which no f-to-m transmission could occur.  It was rolling enrollment so at the early conclusion some men had been involved for only a few months.

ALL the studies shared the design drawbacks discussed at Circumstitions.  To me the biggest flaws are:

- NOT random.  Every subject sought circumcision for cultural reasons and so was perhaps pre-disposed to find value to it or change to more mature behavior or status after the cut.  Subjects were also paid, so there was a potential desire to please benefactors/researchers involved. 

- NOT blinded at all.  Cut men had multiple follow-up visits at the clinic with safe-sex counseling and arm-loads of condoms each time. 

- MANY more subjects lost to follow up than were infected, in either treatment or control groups.  How can we assume these men were infected at the same rate as the remaining subjects?  They might have all gotten sick and sought help at a real hospital.  Or perhaps their circumcisions absessed and they sought treatment elsewhere. 

- NO concordance with the real world.  Half of African nations have higher HIV incidence among the circumcised. 

- REFUTED by the researchers themselves: In 2009 Wawer / Gray reported that the Ugandan men they circumcised were 50% MORE likely to infect their female partners with deadly HIV.  The M-to-F effect washes out any F-to-M benefit, but the WHO has not called for more study to confirm this finding (which is good since it was an immoral Tuskegee-esque atrocity).  In 2010 Bailey followed up and confirmed that men in Kisumu, Kenya were no less likely to be HIV positive if circumcised. 

- IRRELEVANT to infants, who don't have sex. 

- RENOUNCED as human rights abuses by recently updated national medical association policies in Australia and Holland.  They say even if true the studies don't justify infant circumcision and don't justify adult circumcision outside of the regions studied.