Talk:HIV/AIDS/Archive 10

This is an archive of past discussions about HIV. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 5

←

Archive 8

→

Circumcision...

User:Jakew seems to be changing references with regards to the effect of circumcision on the transmission of AIDS. Circumcision seems to be a particular hobby-horse for him, from a look at his talk page. This is probably one area we should check into. --Robert Merkel 02:50, 24 Jun 2005 (UTC)

You are correct. The original link was to an anti-circumcision propaganda web site. I changed this to the most recent systematic review on the subject of circumcision and HIV. User:DanP, who you will note also has a bee in his bonnet on the subject, wanted to include a flawed meta-analysis by Van Howe (an anti-circumcision activist). Since the systematic review I mentioned does reference and discuss Van Howe, and since Van Howe's paper had such serious flaws,[1] [2] I do not think it appropriate to direct readers there, since it would only mislead. You may note that the Cochrane Library review I provided is not especially pro-circumcision (I could have linked to Weiss et al if I wanted). However, it is recent and to my knowledge nobody has made any objection to it. - Jakew 10:51, 24 Jun 2005 (UTC)

I am well aware of your dislike for the Van Howe study. Nonetheless, it does balance the article. There is simply no way you can defend having the WSJ reference in the article, leave out Van Howe, and claim this article is NPOV all at the same time. We are presenting fair and balanced perspectives. DanP 14:19, 16 July 2005 (UTC)

Jakew, that is not a bad edit. But again you've slanted the article in favor of your view. You should at least remove the word "flawed" which the article does not qualify as to who says so, and you should also take out the bit about "suggests", etc. unless you want to apply it equally to the Africa studies, as well as the studies on your site. Just try to balance it, would you, please? DanP 01:41, 17 July 2005 (UTC)

DanP, I'm simply reporting the facts. I've changed 'flawed' to 'methodologically unsound', since that is what was shown. I can't justifiably remove 'suggests', because I have just searched Van Howe for the word 'bias'. It appears three times. In each case, it is followed within two words by 'may'. Van Howe is raising the possibility that they might be biased - 'suggests' seems entirely appropriate. 'Found' would be more suitable for conclusions based upon evidence.

Balance is all very well, but it shouldn't prevent us from being honest. If we were writing an article on gravity, we have little choice but to 'slant the article' in favour of its existence, unless there is credible evidence to the contrary. Sorry. - Jakew 11:57, 17 July 2005 (UTC)

If the Van Howe study is characterized as 'methodolodgically unsound' on the basis you highlight, then the Africa studies are 'methodolodgically unsound' on the basis Van Howe explains. You must do more than report facts, but report them with balance and from primary or secondary sources You can't use your own insight (or some say delusion) to create entirely new information in Wikipedia, that is against the policy on original research. I cannot follow your reasoning at all on how 'may' implies 'suggests'. If one says "smoking may cause cancer", that does not mean the studies behind it merely suggest or hypothesize such a thing without merit. We should report what that the studies "report" or "claim" some connection, not "suggest". Even "suggest" is what you insist upon, the WSJ ought to be so characterized -- otherwise you have editorial slant in your vocabulary. Gravity is entirely irrelevant here, as the topic is circumcision and its effect (both ways) on HIV. Neither causation nor prevention can be given a preference. If it's some popular acceptance of MGM you're getting at to equate that with gravity somehow, remember your side is outnumbered ten-to-one in terms of whether circumcision is believed to be the right choice. Either way, that is not how science is done.

There are obvious methodological flaws in the WSJ-reported study and some of the others:

No strictly double-blind studies (or even any single-blind actually)
No real control group, sham controlled experiments are needed for any meaningful analysis
A report of causation, when they are only detecting correllation
Experiment was never actually completed as planned

I'm sure you will dispute these, but that's OK, I don't think they belong in the article. We can disagree and still find balance. I will change a few sentences to get rid of the slant and let me know if it's OK. DanP 13:19, 17 July 2005 (UTC)

Ok, I've made some changes to your text. Most are ok, but I've restored details of study types. The fact that Van Howe reviewed studies of different design is relevant, as is the distinction between meta-analyses and original studies. I have changed the wording slightly regarding bias. Since he claimed 'may', so do we. I trust you'll agree that this fairly represents his views. I have also commented out your remarks about unknown 'others' who have supported the study. Go ahead and restore it with a reference. - Jakew 14:00, 17 July 2005 (UTC)

The offending tables were removed, so ignore my message below. Wanyonyi 10:44, 16 July 2005 (UTC)

HIV Aids prevalence in non-circumcising AFRICAN countries

Does anyone have statistics on HIV Aids prevalence in non-circumcising AFRICAN countries? The table presented in this section, comparing circumcising and non-circumcising countries, is flawed for several reasons:

The tables compare countries in two completely different locations and socio-economic conditions. It is unrealisitic and bias-inducing to compare zimbabwe and Denmark, for example.
There is no comparison of non-circumcising African countries against circumcising non-African ones.

A comparison of countries such as, say, Kenya (circumcising) and Uganda (largely non-circumcising), or of tribal groups such as Bukusu (circumcising) and Zulu (non-circumcising) will usually indicate higher prevalence in non-circumcising groups. I believe the presentation of the statistics on this page is self-fulfilling, biased and therefore unreliable and unfit for publication in the current format.

See also:

etc Wanyonyi 10:31, 16 July 2005 (UTC)

Agree. I've already stated this (see immediately below), but the person concerned ignored it. There's a good table of African data here.- Jakew 10:50, 16 July 2005 (UTC)

DanP, I've reverted your changes, for the following reasons:

You deleted useful information about the most recent research.
You included a discredited study by Van Howe. You furthermore misrepresented Van Howe's study, because he addressed a different style of research (this study is experimental, while all the ones he addressed were observational).
I have retained your addition about ritual circumcision.
You misrepresent the World Health Organisation, who report no such thing. That is your interpretation.
You include a carefully selected set of countries to support your claim. With the exception of the USA, you're comparing third-world nations with first-world. This is ludicrous.

- Jakew 10:43, 14 July 2005 (UTC)

Don't mean to be a stickler, but the site to allAfrica.com has not mention of any of the details of the study on ritual circumcision. How empirical is this? With a article title like: Ritual Circumcision Likely Contributing to HIV Spread - it makes it seem like they ain't too sure, you know? It is also a reall short article. JoeSmack (talk) 12:58, July 14, 2005 (UTC)

I rewrote this section slightly to emphasize that it's only talking about female-to-male transmission; otherwise the "70 percent" figure is quite misleading. I also added a statement that I think is uncontroversial, but that I can't find a source for right now: Female-to-male sexual transmission is an uncommon route of HIV infection in the Western world, but somewhat more common in Africa. Could someone expand on that? ←Hob 18:24, July 14, 2005 (UTC)

Good rewrite. My understanding is that f-to-m HIV transmission is more common in Africa (and India), simply because such a large proportion of females are infected. That's only my guess. As for JoeSmack's comments about, I have to agree. Looking in the literature, I found some confirmation (eg here), but it's another speculative piece. Intuitively, if you're circumcising tens of boys with the same knife, and one or more is HIV+, some infections are obviously going to happen, but we should probably reword to something like 'it has been suggested that'. DanP: any objections? - Jakew 18:53, 14 July 2005 (UTC) - not here, sounds good to me. JoeSmack (talk) 18:55, July 14, 2005 (UTC)

Indeed, it's a horrible representation to which the paragraph was reverted. Jakew, I would ask you to review what you're defending and ask yourself which portion of the research is the data, and which is the interpretation (sound familiar?). Even worse, the depiction of the "benefit" goes beyond even what the WSJ reports, claiming that circumcision status itself was the cause of a statistical difference, and not the occurence of circumcision of adult men studied within the time frame of the research data. The latter is defensible, the former was never studied. Also, claims of unethical behavior were projected by the researchers, not claimed by WSJ itself. The paragraph you've reverted to is plainly false. This inherent misrepresentation of fact should not stand in Wikipedia. DanP 22:14, 15 July 2005 (UTC)

DanP, do you have the foggiest idea of what the study did? It was experimental. 3000 men volunteered. Half were chosen at random to be circumcised. All men received safe sex counselling. The men circumcised were much less likely to develop HIV. This isn't some sample of men who happened to have HIV at the time and you can make excuses about them being Muslims or something. It was an experiment. They tried circumcision. It worked. The article does not claim that the WSJ said it was unethical. You've ignored my other remarks. I've reverted your changes. - Jakew 10:50, 16 July 2005 (UTC)

I would propose that you simply cite those tables unless they can be made...well...less ug-mo. JoeSmack (talk) 22:41, July 15, 2005 (UTC)

JakeW, here is what you've reverted to:

Preliminary results in a study on the prevention of HIV transmission via circumcision were released July 6, 2005
in the Wall Street Journal, and showed that male circumcision reduced the risk of HIV transmission from women
to men by 70 percent -- a level of protection far better than the 30 percent risk reduction set as a target for an AIDS
vaccine. According to the newspaper account, the study under way in Orange Farm township, South Africa, was stopped because
the results were so favorable: it was deemed unethical to continue the trial after an early peek at data showed that the
uncircumcised men were so much more likely to become infected.[3]
Female-to-male sexual transmission is an uncommon route of HIV infection in the Western world, but somewhat more common in
Africa.

The title refers to circumcision, and that blanket terminology implies the study applied to infant circumcision.
"far better" is POV, so is "favorable" regardless how we feel about specific results.
The use of the vaccine target clearly implies the two are mutually exclusive, suggesting the vaccine is an inferior choice or that it is similarly limited to heterosexual female-to-male pathway. This is a plain misrepresentation.
"According to ... was stopped because" is a claim of what the WSJ says, you should know better than to imply this kind of thing. The ethical matter was an evaluation of the researcher, not the WSJ. Please use accurate facts.
The fact that the study was halted is factual data. The reasons for doing so are interpretation from the research, not scientific method. You know this by now, and you have pointed out this exact distinction before.[4]
There is also no need to specify the exact date of publication in the intro paragraph. This can be linked without the clutter.

Please try to combine your ideas with these instead of a full revert every time. DanP 11:09, 16 July 2005 (UTC)

Not relevant to prevention of AIDS

Interesting though this topic may be is is NOT a method of preventing AIDS - as the cited references make very clear. In fact, circumcism may be the cause of much AIDS in Africa as the cited references also make clear. At best this is controversial research. Sci guy 16:32, 18 July 2005 (UTC)

You're reading too much into the controversy, my friend. Circumcision itself is highly controversial, and anti-circ activists try to confuse just about every issue they can. I've reorganised the section to give a better idea of the chronological order of developments. - Jakew 20:55, 18 July 2005 (UTC)

The issue, my friend, is that Wikipedia is not the place to publish original research or breaking news. A Wall Street Journal news story about research that was rejected for publication by Lancet is not a suitable source of facts for an encylopedia. After this original research has been published you can link to it. Sci guy 02:37, 20 July 2005 (UTC)

The research is not original to Wikipedia. It has been published elsewhere, though certainly I admit it has not yet had the benefit of peer review. Nevertheless, the article does not claim it to be fact - it only states what the article says. Please stop censoring information. - Jakew 10:38, 20 July 2005 (UTC)

A story in a newspaper is not publication. If we reported every newspaper story about AIDS, the AIDS article would be a confused mess. Sci guy 03:47, 21 July 2005 (UTC)

Circumcision controversy section 18 July 2005

Many studies have explored the suggestion that uncircumcised men may be more vulnerable to HIV infection due to a specific type of cell found in the foreskin tissue.

A review and meta-analysis investigating earlier studies of different design claims this reduced risk may have been repeatedly been reported with bias, and concluded that there is no scientific basis for circumcision as a prevention for HIV infection in humans (International Journal of STD & AIDS 1999) [5]. Some critics have claimed that that study inappropriately combines data from differing studies.[6][7] Another review and meta-analysis reported a protective effect of circumcision among adult men in Africa.[8] A 2003 Cochrane review found a strong epidemiological association between male circumcision and prevention of HIV, but recommended waiting for the outcome of randomised controlled trials.[9]

Results of the first randomised controlled trial to investigate the issue were described in the Wall Street Journal[10] on the prevention of HIV transmission claimed that adult men circumcised during the study had a reduced the risk of contracting HIV. Limiting the analysis exclusively to the female-to-male pathway of infection, a reduction of 70 percent was reported. According to the newspaper account, the study in South Africa was stopped by the data and safety monitoring board for ethical concerns. Female-to-male sexual transmission is an uncommon route of HIV infection in the Western world, but is more common in Africa.

The researchers suggested that uncircumcised men may be more vulnerable to HIV infection due to a specific type of cell found in the foreskin tissue, a suggestion supported by other research.[11] Other researchers have suggested that these risks have more to do with hygiene practices, or with the presence of other sexually transmitted infections causing breaks in the skin.[12]

It has also been reported, however, that HIV infections may have been spread in Africa by ritual circumcision.[13]

The latest

Here it is, off my The Body newsletter:

http://www.thebody.com/kaiser/2005/jul27_05/circumcision_hiv.html

Thanks. WHO/UNAIDS statement here. - Jakew 17:58, 28 July 2005 (UTC)