Reader Comments

Post a new comment on this article

Wild extrapolation

Posted by plosmedicine on 30 Mar 2009 at 23:57 GMT

Author: Hugh Young
Position: editor/publisher
Submitted Date: July 15, 2006
Published Date: July 19, 2006
This comment was originally posted as a “Reader Response” on the publication date indicated above. All Reader Responses are now available as comments.

Already the conclusion from the paper of Williams et al. (1) has been spread around the world with headlines like "Circumcision could save millions - WHO" (2).

Yet it is no more than an extrapolation from the Orange Farm study (3). That study relies on 29 circumcised men who did not contract HIV within an 18 month period, compared with 20 circumcised men who did. So each of those 29 men has been extrapolated to more than 125,000 infections and 93,000 deaths prevented - an extraordinary burden of surrogacy. Any faults or inaccuracies - including the statistical inaccuracies imposed by the law of small numbers - are magnified by those figures.

The paper silently assumes:

* circumcision is cost-free and risk-free

* all circumcisions are equally effective

* circumcision has no effect on sexual behaviour

* campaigns to promote circumcision will have no effect on other AIDS-prevention measures.

All of these assumptions are false.

In a knowledge-base where many men think having sex with a virgin will cure AIDS, it is going to be impossible to distinguish this quite slight epidemiologically protective effect from a major protective effect for the individual, especially after a painful and marking operation to the genitals that alters the experience of sexual intercourse. Circumcision en masse will have nothing like the educational side-effect of these circumcisions in their experimental setting. "I don't need a condom, I've been circumcised" will become a commonplace.

Whatever protective effect circumcision may have is going to be totally overwhelmed by irresponsible publicity and ill-informed people's naturally wishful thinking.

Wishful thinking may already have influenced the test subjects (people are not 100% truthful in reporting their sexual behaviour, for example) and the experimenters. The trials were not and could not be double blind, and as I pointed out (4), there was no placebo operation for the control group. Auvert et al. (5) claim this would not have been ethically approved or acceptable to experimential subjects, but these are limitations nevertheless.

A further question arises, about the involvement of the World Health Organisation: the headlines and news stories imply that all of its authority has swung in behind circumcision, when it seems only two of its staff were involved.


1. Williams BG, Lloyd-Smith JO, Gouws E, Hankins C, Getz WM, et al. (2006) The Potential Impact of Male Circumcision on HIV in Sub-Saharan Africa. PLoS Med 3(7): e262

2. - Circumcision could save millions - WHO. Dominion Post, Wellington, New Zealand, July 12, 2006.

3. Auvert B, Taljaard D, Lagarde E, Sobngwi-Tambekou J, Sitta R et al. (2005) Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: The ANRS 1265 trial. PLoS Med 2: e298 DOI: 10.1371/journal.pmed.0020298.

4. Young, Hugh (2005) Two groups not on all fours. PLoS Med 2: http://medicine.plosjourn...

5. Auvert, Bertran et al. (2006) Corresponding Authors' reply PLoS Med 2: http://medicine.plosjourn...

No competing interests declared.