Talk:Circumcision

Citing Cochrane

A concern has been expressed that the "strong evidence" sentence shouldn't be presented in Wikipedia's voice unattributed. However the attempts to address this concern have ended up mis-citing or mis-characterizing Cochrane. The Cochrane review isn't just "a study," it's a meta-analysis of several studies, with heavy statistical analysis and corrections for confounding factors. Per WP:MEDRS and other Wikipedia policy, this sort of meta-analysis of large bodies of data is "Wikipedia gold"--from WP:MEDRS: "Ideal sources for such content includes general or systematic reviews published in reputable medical journals, academic and professional books written by experts in the relevant field and from a respected publisher, and medical guidelines or position statements from nationally or internationally recognised expert bodies." We have exactly this. In my editorial judgment, the cites we have are strong enough to make a statement about the evidence in Wikipedia's voice. I feel like if this article were ibuprofen instead of circumcision, and we had the same kinds of sources, the article would say "Ibuprofen reduces pain by 40%" or whatever, without qualification or controversy, or even the need to say "Meta-analyses of studies show that there is strong indicates that indicates ibuprofen reduces pain by 40%." So for at least the time being, I am offering a compromise that attributes the results to Cochrane, and accurately characterizes both that kind of study Cochrane is, and the results. There is a second issue of discussing criticism of the Cochrane study in the article, and I'll open a separate Talk page topic about that. Zad68 14:39, 27 May 2012 (UTC)

experimental research results for Ibuprofen have been replicated and confirmed everywhere in the world, acorss all boundaries cultures peoples; the HIV benefit of circumcision has been subjected to no rigorous skepticism and experiemtnal testing. And, yes, we're still waiting for your explanation about the initial editorial motivation to include these tribal studies as final ande absolute fact for the introduction. Robert B19 (talk) 18:33, 27 May 2012 (UTC)

Robert, please review WP:MEDRS, specifically where it says "Ideal sources for biomedical material include general or systematic reviews in reliable, third-party, published sources, such as reputable medical journals, widely recognised standard textbooks written by experts in a field, or medical guidelines and position statements from nationally or internationally recognised expert bodies." The Cochrane review and WHO statements meet Wikipedia's highest standards as references. Requiring further studies that re-confirm what the Cochrane review and WHO already say is outside Wikipedia guidelines, and would fall under a "Moving the goalposts" style of argumentative fallacy.

For what it's worth, I looked at aspirin. (The article aspirin is a WP:GOOD article and ibuprofen is not, so I thought it would be a better demonstration of Wikipedia's standards at work.) Aspirin says "In general, aspirin works well for dull, throbbing pain;..." and that is attributed to one source--a single journal article by one author. The article doesn't say "In one study, Zbigniew Gaciong (2003) found aspirin generally to work well for dull, throbbing pain..." it simply says "In general, aspirin works well for dull, throbbing pain" and the sourcing for that statement is nowhere near as high-quality as what we have for circumcision. Based on this, wouldn't you agree that the Cochrane and WHO findings meet Wikipedia requirements for unqualified inclusion, written in Wikipedia's voice? Zad68 16:14, 31 May 2012 (UTC)

Agree with Zad68. This seems like a reasonable compromise, as Cochrane reviews are widely held as high quality reviews. Yobol (talk) 21:00, 27 May 2012 (UTC)

Its far from a reasonable compromise. Try again. Pass a Method talk 22:29, 27 May 2012 (UTC)

Perhaps you could explain what your objections are, and without the snark. Yobol (talk) 22:37, 27 May 2012 (UTC)

The reason i hesitate to make a new compromises is that you guys sometimes make it even less neutral than the previous ones. In order for us to reach an agreement, Yobol, Zad68 and Jakew need to ask themselves why so many editors have long complained about the biased state of the lede. I have edited hundreds of articles and the circumcision lede yesterday was by far the most non-neutral article i've seen. It is still non-neural now. Anyody so myopic to be unable to have seen the extreme bias in yesterday's lede I am unable to have a conversation with since we are obviously living on different planets. Pass a Method talk 23:00, 27 May 2012 (UTC)

So instead of addressing directly the attempt at compromise, you declare everyone who has ever disagreed with your position as the problem, and refuse to engage in the conversation? Why exactly are you here on the talk page if you aren't willing to discuss the article? Yobol (talk) 23:05, 27 May 2012 (UTC)

Im willing to discuss the article, but seeing how previous discussions have gone, im not optimistic. I sort of feel like trying to compromise with people who defended yesterdays version is like trying to convince Joseph McCarthy to support free trade. It is a lot more difficult than you think. Pass a Method talk 23:10, 27 May 2012 (UTC)

So discuss. What is wrong with the wording as proposed? Yobol (talk) 23:12, 27 May 2012 (UTC)

I have some reservations about it, but I think Zad68's edit is an acceptable compromise. Jakew (talk) 07:12, 28 May 2012 (UTC)

Typically when a source is a good as Cochrane we simply state the facts "In Africa circumcision reduces the risk of HIV infection in heterosexual men by 38-66%". However the compromise is reasonable.Doc James (talk · contribs · email) 15:36, 31 May 2012 (UTC)

My issue with this revision is that we are still taking a study conducted in Africa and extrapolating the results to apply to heterosexual men everywhere. The lead still states that the analysis "... indicates circumcision reduces the risk of HIV infection in heterosexual men by 38-66%" (emph. mine). We know the RCTs surveyed many high-risk individuals. The authors of the RCTs seemed to acknowledge this as well. To my knowledge, most of the rest of the world is not at high risk of HIV transmission. Prior to this discussion, the lead stated for over half a year that "circumcision has been shown to reduce the risk of acquiring HIV in heterosexual populations that are at high risk". I'd like to recap the conclusions made by two of the three authors of the African RCTs:

• In PMID 17321310, the authors of the Kenyan RCTs state in its abstract that "Male circumcision significantly reduces the risk of HIV acquisition in young men in Africa".
• The authors of the African RCT stated the following: "This study has some limitations. It was conducted in one area in sub-Saharan Africa and, therefore, may not be generalizable to other places. Nevertheless, because of the similar route of transmission of HIV in sub-Saharan Africa and because observational studies from various areas of sub-Saharan Africa have shown an association between HIV status and MC, the result of this trial is applicable to all of sub-Saharan Africa with some degree of confidence."

With Wikipedia:Verifiability in mind, our lead seems to be inconsistent with the conclusions drawn from 2 of the 3 authors of the RCTs. FactoidDroid (talk) 03:07, 1 June 2012 (UTC)

We discussed this at length in Talk:Circumcision/Archive 72#RFC, didn't we? Also, even if it were a valid objection, I don't see how it differentiates between the three versions (ie., the long-standing "Strong evidence from Africa" version, the "One analysis of studies done in Africa" version, and the proposed compromise "A Cochrane meta-analysis" version), as all of these refer to heterosexual men. Jakew (talk) 09:40, 1 June 2012 (UTC)

An important distinction that could be made in the lead is that is has been shown to decrease female-to-male HIV transmission in African men. This is supported by the original authors of the studies, as I've shown, but the lead is currently far more generalized. FactoidDroid (talk) 02:50, 4 June 2012 (UTC)

As I said, wasn't that the subject we discussed in Talk:Circumcision/Archive 72#RFC? Jakew (talk) 07:22, 4 June 2012 (UTC)

(Unindenting) It looks as though we have rough consensus above in favour of this compromise. There are only two objections, one of which is a comment on contributors rather than content, and the other (which revisits the "African men" suggestion) applies equally to the other versions, and therefore is not a reason to reject this compromise. Jakew (talk) 09:51, 6 June 2012 (UTC)

Agree. Zad68 12:19, 6 June 2012 (UTC)

Passamethod, Robert, NeilN, and I reverted Zad's compromise in favor of the current compromise. I still believe if we are going to have this detail of HIV in the lead it must be balanced with its own prevalent well sourced criticsm. Garycompugeek (talk) 13:35, 6 June 2012 (UTC)

Incorrect, Gary. Only Robert B19 reverted Zad68's compromise. See [1]. Jakew (talk) 13:38, 6 June 2012 (UTC)

The page was protected after Robert reverted Zad's compromise which really isn't a compromise at all. I'll illustrate the changes..

original

"Strong evidence from Africa indicates that circumcision reduces the risk of HIV infection in heterosexual men by 38-66%,"

2nd version

"One analysis of studies done in Africa indicates that circumcision reduces the risk of HIV infection in heterosexual men by 38-66%,"

3rd version

"A Cochrane meta-analysis of three large randomized controlled trials from Africa has shown that there is strong evidence that circumcision reduces the risk of HIV infection in heterosexual men by 38-66%,"

The only thing Zad's compromise accomplishes is to site Cochrane... The edit war was about removing "strong". Garycompugeek (talk) 18:08, 6 June 2012 (UTC)

I might agree with the Zad compromise if it did not have so much detail. Why not "a meta-analysis" without mentioning "Cochrane" or "three large randomized controlled trials" part? Also, why not a brief mention of reviews criticising the RCT's? Pass a Method talk 18:27, 6 June 2012 (UTC)

Recall that in the discussion taking place at the time, the main objection was that "Wikipedia should not be taking the study's conclusions as its own" (this was clearly why the "one analysis" wording was introduced: so as to be able to attribute the conclusion to a particular source). But those objecting to the "one analysis" wording noted that it failed to do justice to the source, which is highly regarded and carries a great deal of weight in evidence-based medicine. Zad68's compromise avoided stating the conclusion in Wikipedia's voice, but gave enough detail about the nature of the source to allow the reader to assign that weight him/herself. "A meta-analysis" doesn't provide enough information to allow the reader to do that. For example, "a meta-analysis" might describe a poorly-regarded meta-analysis of a handful of small, low-quality observational studies. Jakew (talk) 18:41, 6 June 2012 (UTC)

Consdering the study has been criticised by notable medical organizations and reviews, i would apprciate if you stopped using words such as "highly regarded" for Cochrane. Pass a Method talk 19:16, 6 June 2012 (UTC)

A handful of papers have criticised the RCTs, but very few have actually criticised the Cochrane review itself. In any case, like it or not, Cochrane reviews are highly regarded. Jakew (talk) 19:24, 6 June 2012 (UTC)

Thats my point. RCT's and other methods in thie African study have been criticized. by notable medical organizations. Please take that into consideration when you make a proposal. Pass a Method talk 19:41, 6 June 2012 (UTC)

We've already discussed the inclusion of criticism on many occasions. In any case, none of the three wordings refer to criticisms, so that isn't an argument for or against the compromise. Jakew (talk) 19:50, 6 June 2012 (UTC)

btw Doc James above has also indicated he is okay with merely mentioning reduced HIV incidinces. Pass a Method talk 19:54, 6 June 2012 (UTC)

My position has been that a sound editorial judgment precludes the hiv studies from the lead. The prhrse one analysis barely passes muster. I don't see the relevant of the hiv studies for english reader. Robert B19 (talk) 19:57, 6 June 2012 (UTC)

(e/c) Pass a Method, that's exactly the value of a Cochrane review. A Cochrane review is a systematic review that starts with a question to answer, reviews all the available information published about that question, selects the best information sources, and performs heavy meta-analysis on them. So even if some sources question a particular RCT, a systematic review like Cochrane will evaluate those questions, and cross-compare the methods and results of all the included RCTs to produce a very reliable result. It is for exactly this reason that Wikipedia policy WP:MEDRS gives the most weight to a systematic review, and even values Cochrane so highly that it says under "Use up-to-date evidence": "Cochrane Library reviews are generally of high-quality and are routinely maintained even if their initial publication dates fall outside the above window," allowing the use of Cochrane reviews for longer than other study results. Honestly, this unwillingness to accept the results of this Cochrane review is beginning to sound like nothing more than WP:IDONTLIKEIT, which is not an acceptable argument against using it. Zad68 19:59, 6 June 2012 (UTC)

Zad, i have never disregarded CXochrane. I am well aware of its merits. However, i am only asking that we balance it in a larger context. WP:MEDSCI also says that we should "make readers aware of controversies that are stated in reliable sources". Pass a Method talk 20:43, 6 June 2012 (UTC)

Do you have any proposals on how we should word the controversies of the African study? Pass a Method talk 21:24, 6 June 2012 (UTC)

Pass a Method, re your statement "i have never disregarded CXochrane. I am well aware of its merits." -- I'm getting mixed messages here, because earlier you also said "i would apprciate if you stopped using words such as "highly regarded" for Cochrane." We need to clarify this--Can we both agree to the statement "According to Wikipedia standards, the Cochrane review is an ideal type of medical reliable source, and its findings should be used in the article." I agree with this statement, do you?

Regarding "Do you have any proposals on how we should word the controversies of the African study" -- What do you mean by "African study"? Do you mean one of the individual RCTs included in the Cochrane review? Or do you mean the Cochrane review itself? (Is there an objection now to using the phrase "Cochrane review" at all?? I sure hope not.) If you mean "Cochrane review", then this subject of including criticism of Cochrane in the article is actually off-topic for this discussion section, which was meant to be about Citing Cochrane--meaning, how exactly should the Cochrane review results (not the criticism of them) be presented in the lead. Rest assured that myself and Jakew and others (I'm sure) agree that criticism of Cochrane is notable and should be covered by Wikipedia, and I actually already raised the topic of including criticism of Cochrane in the article--see Talk:Circumcision/Archive_72#Criticism_of_Cochrane. Jakew pointed out that the article already includes a mention of criticism of Cochrane and suggested expanding medical analysis of circumcision. You didn't contribute to that discussion, maybe you missed it? If you'd like to discuss that, would you please un-archive that thread and bring it to the top Talk page. For clarity, I'm trying very hard to keep the separate discussions in their appropriate sections. Zad68 02:05, 7 June 2012 (UTC)

Zad, what i meant by that perceived contradiction is that Jake merely intends to mention the African study without mentioning those opposing the study. I believe Jake is giving undue emphasis on one journal and ignoring the dozens of others. The way Jake interprets wikipedia guidelines you almost get the impression that wikipedia policies state "we only use Cochrane. Anything contradicting Cochrane has to be disregarded, deleted and/or ignored." Pass a Method talk 02:28, 7 June 2012 (UTC)

In the interests of clarity, my position is that discussing critics in the lead would give them undue weight, since even a few words would represent a significant fraction of the total HIV material in the lead. However, I see nothing wrong with brief discussion in the body of the article (as is already the case), or lengthier discussion in one of the more detailed articles. As Zad68 has commented, however, this discussion is off-topic for this thread. Jakew (talk) 07:35, 7 June 2012 (UTC)

Jakew, I agree with you here. Zad68 19:45, 7 June 2012 (UTC)

Pass a Method, again, there is consensus from editors here, including myself and Jake, to mention the criticism of Cochrane in the article. Jake and I agree that, per WP:LEAD and WP:UNDUE, the correct placement of the criticism of Cochrane would be a sentence in the body mentioning it, and possibly expanding on it in medical analysis of circumcision. And, again, the article already mentions the criticism of Cochrane, in the body. It appears, based on your contributions here and your edits to the article, that you are generally in agreement with the ideas that: 1) Cochrane should be used and cited, and 2) Cochrane could be cited in the lead with a sentence like, "A Cochrane meta-analysis of three large randomized controlled trials from Africa has shown that there is strong evidence that circumcision reduces the risk of HIV infection in heterosexual men by 38-66%." This is good, because we actually need get to a consensus on these topics so that the article can remain stable, and we can dispense with the slow (or fast...) edit-warring and other stuff that lead to things like the full-protection and 1RR proposals. With that, I'll un-archive the "Criticism of Cochrane" thread from the archive, and let's continue the discussion of the topic there. Zad68 19:45, 7 June 2012 (UTC)

You can't get aroundthemisleading wording that makes the african tribeal studies applicable to men elwhere. Robert B19 (talk) 22:37, 7 June 2012 (UTC)

There is nothing "misleading" in the source. We should stick to the wording of the source, which discusses risk in heterosexual men with no qualification. Yobol (talk) 23:16, 7 June 2012 (UTC)

qualifies it quite a bit Yobol"The review supported previous review fingings thatt the results from existing obessevational studies showed a strong epidemiological associaltion between male cirucmsiiion and prevention of HIv, espeically among high-riskgroups." — Preceding unsigned comment added by Robert B19 (talk • contribs) 22:00, 9 June 2012 (UTC)

The wording of the source without a qualifier runs contrary to the conclusions stated by two of the three authors of the RCTs (as I've pointed out in PMID 17321310 and this link). Perhaps I'm having trouble reconciling this with Wikipedia:Verifiability. FactoidDroid (talk) 06:57, 14 June 2012 (UTC)

In terms of WP:V, there's a very important question: which source do we cite, and does that source support the qualifier? Jakew (talk) 07:33, 14 June 2012 (UTC)

Do Van Howe & Storm's criticism of Cochrane discredit or cast enough doubt to rule out its use?

A few editors have brought up Van Howe & Storm's criticism of Cochrane, and feel that their criticism discredits or casts enough doubt on Cochrane such that, per Wikipedia policy, it shouldn't be used. I'd like to hear the arguments why. On the outset, the credentials behind the Cochrane review appear to far outweigh Van Howe and Storm; Van Howe and Storm's analysis has itself been criticized by peers who have found enough error in Van Howe and Storm's criticism to cast at least equal doubt on their criticism; and Wikipedia policy weights well-respected meta-reviews much more highly than individual journal articles. Would those who are making the argument that Van Howe & Storm's criticism rules out the use of Cochrane please address this here. Zad68 18:11, 31 May 2012 (UTC)

The source for that article the "Journal of Public Health in Africa" is an open sourced journal that is not MEDLINE indexed, with a dubious looking editorial team (all of one person that is back in the main publishing house, PagePRESS). I doubt this qualifies as a WP:MEDRS, and would give zero weight to it. Even if it were to be used, it should be given considerably less weight than a Cochrane review (one of the highest quality MEDRS sources available), and certainly cannot be used to disqualify the Cochrane source. Yobol (talk) 18:16, 31 May 2012 (UTC)

Yobol, although I'm inclined to agree with you, Cochrane has (I believe) been in the article supporting content for quite some time now. It's a new proposal from some editors to remove the source from the article; per WP:BURDEN it's up to them to make a case for removing it to a sufficient degree that it should be removed per consensus, so I'd like to hear the arguments. Zad68 18:21, 31 May 2012 (UTC)

The text by Van Howe and Storm is not the only one criticizing the analysis. PMID 22320006, published in the Australian Journal of Law and Medicine, seems to take issue with many aspects of the analysis, stating "the trials were compromised by inadequate equipoise; selection bias; inadequate blinding; problematic randomisation; trials stopped early with exaggerated treatment effects; and not investigating non-sexual transmission." PMID 21973253, published in the Australian and New Zealand Journal of Public Health, criticizes the Cochrane analysis with respect to its application in Australia. The conclusion of the text reads: "Our conclusion is that such proposals ignore doubts about the robustness of the evidence from the African random-controlled trials as to the protective effect of circumcision and the practical value of circumcision as a means of HIV control". The Royal Dutch Medical Association has also scrutinized the studies, stating: "That the relationship between circumcision and transmission of HIV is at the very least unclear is illustrated by the fact that the US combines a high prevalence of STDs and HIV infections with a high percentage of routine circumcisions. The Dutch situation is precisely the reverse: a low prevalence of HIV/AIDS combined with a relatively low number of circumcisions. As such, behavioural factors appear to play a far more important role than whether or not one has a foreskin." The very title of PMID 20965388, published in The American Journal of Preventative Medicine, is "Male circumcision and HIV prevention insufficient evidence and neglected external validity".

At the very least, there seems to be merit to at least integrating into the lead that the analysis have been in scrutinized by valid sources such as those indexed by pubmed that I've linked to above. FactoidDroid (talk) 04:05, 1 June 2012 (UTC)

Per WP:UNDUE, the number of critical sources is less important than the proportion ("Wikipedia aims to present competing views in proportion to their representation in reliable sources on the subject"). Five critical sources in ten, say, is a significant number, and (all things being equal) would suggest similar weight. Five critical sources in fifty is much less significant. In User:Jakew/reviewsumm I sampled sixty of the most recent secondary sources in PubMed. Only two were negative towards circumcision's partial protection against female-to-male transmission. Jakew (talk) 07:06, 1 June 2012 (UTC)

Jake, i think you may have a tilted understanding of WP:UNDUE. I believe the most important notion is the following question. How do we find the right balance between the conflicting references? Although our conclusion to what that balance is may differ. Pass a Method talk 16:41, 3 June 2012 (UTC)

Do you agree that WP:UNDUE states "Wikipedia aims to present competing views in proportion to their representation in reliable sources on the subject", PassaMethod? Jakew (talk) 16:59, 3 June 2012 (UTC)

Yes i do, which bringe me to FactoidDroid's point above in the last sentence about valid scrutiny. Pass a Method talk 17:26, 3 June 2012 (UTC)

Since you agree that the goal is to present views in proportion to their representation in reliable sources, would you explain why determining the proportion to which they're represented in reliable sources is inconsistent with WP:UNDUE? Jakew (talk) 17:35, 3 June 2012 (UTC)

I did not say that it is. Pass a Method talk 18:04, 3 June 2012 (UTC)

Jakew, I imagine it would help if you mentioned how many of the sixty secondary sources had a discernible viewpoint one way or the other. That would help us put the proportion in better perspective. FactoidDroid (talk) 02:46, 4 June 2012 (UTC)

Thirteen are marked 'unclear', leaving thirty forty seven. Jakew (talk) 11:56, 4 June 2012 (UTC)

Jakew, I believe if you were to sample articles talking about circumcision anywhere but in Africa, you would find that very few actually address the validity of the African trials with respect to other continents; those few that do, are negative. The weight, then, is not 5 articles out of 50, but 5 articles out of maybe 7.

To further comment on the criticisms of the Cochrane review, these criticisms should focus on the applicability of the meta-analysis to other cultures, not on the methodology of the meta-analysis itself with regards to reducing HIV in Africa, since many articles reference Cochrane in reducing HIV in Africa. Rip-Saw (talk) 23:54, 7 June 2012 (UTC)

Rip-Saw, in this comment you mention that you had "actually read the article in question" and, therefore, it was "clear" to you that the Cochrane authors did not intend to generalize their conclusion to outside Africa. I have also read the article, and do not see what you were talking about. Could you please point out what part of the Cochrane study shows the Cochrane authors did not intend to generalize their conclusion? I would certainly hope not to be a part of "grossly irresponsible scientific reporting". Thanks. Yobol (talk) 00:27, 8 June 2012 (UTC)

Rip-Saw, I'm afraid your comments don't make a lot of sense. First of all, the scope of the sentences we're discussing is not restricted to non-African environments. Consequently, there is no logical basis for assessing due weight as you propose. Secondly, in a hypothetical situation in which such a method were appropriate, your method still wouldn't make sense because you haven't provided any evidence in support of your figures. Jakew (talk) 09:42, 8 June 2012 (UTC)

Yobol, I actually said it was clear that the authors intended the results apply to Africa. What is not clear is if they intended the results apply to other cultures outside the ones they studied. In science, it is assumed you do not come to conclusions about something without having first studied it. Since the authors never make any mention of cultures outside of Africa, I do not believe they intended to make such a scientific blunder. Their words are being misunderstood and taken out of context.

Jakew, the vast majority of articles that make reference to the Cochrane review are articles about Africa. In that respect, when it comes to Africa, the vast majority of literature supports the Cochrane reviews, and thus would be undue to call into question the validity of the review. Outside of the scope of Africa, the literature is split, with a substantial percent questioning the Cochrane review's external validity. This means it is perfectly OK to discuss the problems with the review in this context. To not do so would be POV. Rip-Saw (talk) 15:40, 8 June 2012 (UTC)

If we find ourselves discussing protective effects by continent, let's work out due weight accordingly. For the time being, however, it doesn't make sense to assess due weight in a different context than that of the sentences involved. Jakew (talk) 15:54, 8 June 2012 (UTC)

There is a conflict inmeaning then. Footnote 16, "Economic Evaluations of Adult Male Circumcision for Prevention of Heterosexual Acquisition of HIV in Men in Sub-Saharan Africa: A Systematic Review", refers to the first part of the sentence, "A Cochrane meta-analysis of studies done in Africa indicates that circumcision reduces the acquisition of HIV by heterosexual men between 38% and 66% over 24 months". It's quite clear that the findings apply only to african or footnote 16 can't be included. else it impliesthat circumcision has been found costeffective for men everywhere, not just afraica which the footnote doesnt support. Robert B19 (talk) 18:28, 8 June 2012 (UTC)

{Outdent}Robert what is the complete first sentence in the abstract at the very beginning of the article referenced by footnote 16? Zad68 19:41, 8 June 2012 (UTC)

@Rip-Saw: I think it takes a very unusual understanding of the words "out of context" to suggest using the words that the authors used, when there is no evidence to the contrary in the document, is somehow incorrect. The authors, at no point, suggest or say otherwise, but you have made it up in your mind you know what they are truly thinking and saying, against the plain reading of their words. Remarkable. Yobol (talk) 01:22, 9 June 2012 (UTC)

The Cochrane Review group on hiv/aid disagrees with you. Safety of task shifting for male medical circumcisionin Africa: a systematic review and meta-analysis "We restricted our reviewto studies done in Africa, as the region where malemedical circumcision is predominately being rolled outfor HIV prevention." http://hiv.cochrane.org/sites/hiv.cochrane.org/files/uploads/Ford_2011.pdf Robert B19 (talk) 22:24, 9 June 2012 (UTC)

There are a number of reviews on the topic of HIV and circumcision, and the one you are linking is a completely different one than the one being used in the lead and is being discussed. Please stay focused on the topic. Yobol (talk) 22:47, 9 June 2012 (UTC)

{Outdent} Proposed revision to lead: "A Cochrane meta-analysis of circumcision experiments done on sexually active men in Africa indicates that circumcision reduces the acquisition of HIV by heterosexual men between 38% and 66% over 24 months" The lead probably misleads the layperson about the experiemnts. The layperson will assume we're desribing a beneit of infant circumcison. Can't we clarify it so there's no possiblity of confustion? Robert B19 (talk) 17:06, 10 June 2012 (UTC)

The first problem there is that the word "circumcision" in "circumcision experiments" is unnecessary. The second is that "experiments" is imprecise: "randomised controlled trials" is more informative, particularly if linked. "Done on" is slightly awkward wording. Finally, "circumcision reduces the acquisition of HIV by heterosexual men by between 38% and 66% over 24 months" is copied verbatim from the source, except that the word "by" is omitted, introducing a grammatical error. See WP:PLAGIARISM for an explanation as to why this is a problem.

To correct these issues, I've rewritten it as follows: "A Cochrane meta-analysis of African randomized controlled trials indicates that circumcision reduces HIV infections among heterosexual men by 38%-66% over a period of 24 months." Jakew (talk) 17:27, 10 June 2012 (UTC)

Criticism of Cochrane

There has been expressed a desire to include some content on the criticism of the Cochrane study and its findings. There are enough sources out there that this topic in and of itself is notable. I would like to open a discussion about what to include, and where. I observe that this article Circumcision is already too long. It is the 1,638th largest of several million articles (putting it easily in the top 0.1% largest articles), and as it currently stands is over 140,000 raw bytes, and so is probably WELL into the Wikipedia recommendation to be divided up--"Readable prose size > 100 KB: Almost certainly should be divided," see WP:SIZERULE. According to Wikipedia guidelines, including WP:UNDUE, the article would be actively made worse by adding a large section about the criticism of this one source of data supporting maybe a dozen sentences in the article. I suggest that a new article be created, something along the lines of Criticism of medical analysis of circumcision, that a one-sentence summary along the lines of "Various sources and studies have criticized the methodology and findings of Cochrane and other studies," along with a "See also" linking to that new article. Thoughts? Zad68 15:10, 27 May 2012 (UTC)

I think it's a good idea to consider where might be a good place to put this material. However, WP:CRITICISM discourages "criticism of" articles, and for good reason, I think, as it's very hard to maintain NPOV with such a scope. I see no reason why this material shouldn't be included in medical analysis of circumcision or circumcision and HIV, where there is plenty of space. A section of circumcision and HIV would probably be the best option. In this article, I'd like to point out that we do already cite criticism in the body ("Some have challenged the validity of the African randomized controlled trials, prompting a number of researchers to question the effectiveness of circumcision as an HIV prevention strategy.[95][96]"); does that wording need to be changed? Jakew (talk) 15:51, 27 May 2012 (UTC)

Unarchived for further discussion. Zad68 19:50, 7 June 2012 (UTC)

I think there is a lot of of merit to creating a page such as "Criticism of medical analysis of circumcision". I would be more than happy to begin generating content for this article. I feel that "Some have challenged the validity of the African randomized controlled trials, prompting a number of researchers to question the effectiveness of circumcision as an HIV prevention strategy" is far too ambiguous for the scope of the criticisms that have been made, such as those that I pointed out in my edit dated 04:05, 1 June 2012 (UTC) here. FactoidDroid (talk) 03:17, 8 June 2012 (UTC)

We don't need more WP:POVFORKs like "Criticisms of..." pages. Expansion into circumcision and HIV would be most logical. Yobol (talk) 03:31, 8 June 2012 (UTC)

I agree with Yobol's and Jakew's reading of the Wikipedia recommendations here, they make a good point that there should not be a separate "Criticism of..." article, due to the likely WP:POVFORK and WP:NPOV issues that would arise. FactoidDroid, are you on board with putting this content in circumcision and HIV? If so, we'd be starting to have consensus for that and could move forward. Zad68 13:23, 8 June 2012 (UTC)

Lead change removing Africa HIV information

The reason why the content was removed is as follows:

1. Most circumcisions are done for non-therapeutic reasons on infants
2. The Africa HIV claims are relevant to adult men in areas with high HIV prevalence among heterosexuals

Therefore:

The Africa HIV claim should not be in the lead since those claims are not relevant to the majority of the reasons for circumcisions. As WP:LEAD states: "material in the lead should reflect its relative importance to the subject". At best, this viewpoint represents a very low level of importance to the subject.

Because this information is of low importance to circumcision and the majority of reasons for circumcision, placing it in the lead violates WP:NPOV.

Furthermore, reliable secondary sources such as http://xa.yimg.com/kq/groups/23477339/1441224426/name/JLM_boyle_hill.pdf have cast serious doubt on the Africa HIV studies. This is all the more reason why this information should be kept out of the lead. It is not clear this is reliable information. The information that I removed from the lead is present in the article already, so I'm not removing facts or attempting to bias the article. The change I made makes the article more neutral as it removes a dubious, not-so-relevant claim from a high-visibility area of the article.

Additionally, my reasons for removal of this information are echoed by other Wikipedia editors. Please see Wikipedia_talk:Manual_of_Style/Lead_section#Clarification_for_scientific.2Fmedical_articles. Crimsoncorvid (talk) 02:51, 14 June 2012 (UTC)

Proposals to remove that particular paragraph has undergone discussion for months (see for example this thread in the archives. It would be more fruitful for everyone involved if you were to read through the archive before propopsing similar changes which have never received consensus for removal. Note that those "other editors" in that thread were not told by you the specifics of this discussion and were only working on a hypothetical scenario which you proposed, and therefore holds no weight in this discussion. Yobol (talk) 03:05, 14 June 2012 (UTC)

Thanks for the pointer to the archives. It seems like most of the editors favor removing this statement from the lead so I am a bit confused. Also, I don't see the points I just brought up addressed directly. Crimsoncorvid (talk) 03:12, 14 June 2012 (UTC)

Perhaps you should read the thread I linked to again, where several editors, including myself, have given their reasons for including a discussion of HIV in the lead. We should not endlessly debate the same topic ad nauseum when they have alredy been discussed. Yobol (talk) 03:17, 14 June 2012 (UTC)

You mentioned that similar changes never reached consensus. We must continue to discuss this issue, obviously. It might be helpful if you could point to where my view on this issue is mistaken. I don't see the flaw in my reasoning and, as I mentioned before, the archived discussion does not directly address the points that I am making here.

I have read WP:DEMOCRACY so I know we're not supposed to "vote" per say, but I see more editors in favor of removing the Africa HIV statements. What exactly do you mean by "consensus" on this topic? Crimsoncorvid (talk) 03:51, 14 June 2012 (UTC)

I will provide a personal view that is expressed too strongly: Consensus is agreement between experienced editors regarding the best resolution of a proposal, where "best" is measured in terms of agreement with existing policies and practices. An "experienced editor" is one who has done significant work in a number of unrelated articles (for example, someone whose sole experience was adding a certain slant to all articles regarding a political event would be regarded as an WP:SPA, and their opinion on Wikipedia's practices would not rate highly). Johnuniq (talk) 07:48, 14 June 2012 (UTC)

Crimsoncorvid, re: "I don't see the flaw in my reasoning and, as I mentioned before, the archived discussion does not directly address the points that I am making here." -- Addressing your points directly: The policy cited ("material in the lead should reflect its relative importance to the subject") is the relevant one, but the flaw in the reasoning is a misunderstanding or misapplication of it. To understand what that bit from WP:LEAD means, you need to read WP:WEIGHT, which is part of Wikipedia core content policy. The relevant sentence is: "Neutrality requires that each article or other page in the mainspace fairly represents all significant viewpoints that have been published by reliable sources, in proportion to the prominence of each viewpoint." (emphasis mine). So the relevant amount of weight and detail in the article to a viewpoint is not in proportion to the prevalence of reasons why the procedure is performed (as you're suggesting), but rather in proportion to its discussion in reliable sources. As a rough test of whether reliable sources give high prominence to discussing HIV in covering circumcision, do this quick exercise: Go to scholar.google.com and search on "circumcision", how many of the search results mention HIV? Answer: About half. So, it has been well-established that reliable sources give high prominence to discussing circumcision's effect on HIV transmission, and that is the reason why it is discussed in the proportion it is in the body of the article, and, therefore, that's the reason why 2 sentences of the lead discuss HIV. Zad68 12:33, 14 June 2012 (UTC)

Can someone point to the original talk page consensus to include the Cochrane reference in the article lead? I can't find it in the archives. AvocadosTheorem (talk) 17:58, 14 June 2012 (UTC)

Per WP:Consensus#No consensus, "In discussions of textual additions or editorial alterations, a lack of consensus results in no change in the article." This particular change (removal) has been discussed for months now, and has never achieved consensus for removal. Yobol (talk) 18:06, 14 June 2012 (UTC)

I'm searching for the Archived Talk page that initially established the consensus to include the Cochrane reference in the article lead. I'd like to read it before weighing in on the current dispute to keep the reference. AvocadosTheorem (talk) 18:47, 14 June 2012 (UTC)

Thanks Yobol but AvocadosTheorem asked where in the archives is consensus for Cochrane referenced inclusion not removal. To be honest there has been mention of HIV in the lead since 2007 from the WHO and the CDC in 2008, which was replaced by the cochrane meta analysis in 2009. Myself and others have lobbied to remove it from the lead (not the article) with little success. I don't believe it has ever enjoyed consensus in the usual sense but it has Jake's blessing which amounts to the same thing in circumcision related articles. Garycompugeek (talk) 18:56, 14 June 2012 (UTC)

I was not involved in this article back then so I wouldn't know where to start looking for a discussion 3 years ago. As things stand, when there is no consensus to change, the default is to keep it how it is. If something has been in the article since 2009, it has consensus, and it is up to those who want it changed (in this case removed) to establish that new consensus to remove it. Yobol (talk) 19:01, 14 June 2012 (UTC)

I fail to see how this is consensus among editors just because the content has been there "for a long time". It is time to re-evaluate. Crimsoncorvid (talk) 06:10, 15 June 2012 (UTC)

It is what it is. Material that has been in the article for 3 years has the benefit of presumption of having consensus. If you want to change that material, the burden is on you to show the consensus has changed. Yobol (talk) 14:09, 15 June 2012 (UTC)

With the exception of the US CDC and NIH, not a single national medical organization has endorsed the Africa HIV studies. Brazil, Australia, and a number of European countries have issued statements rejecting these claims. The WHO and Cochrane studies have been called into question. I do not see consensus. Counting up medical studies is not consensus as that is proportional to the research funding. When an organization is trying to promote a point-of-view, it will fund studies in this area to generate support. Crimsoncorvid (talk) 06:10, 15 June 2012 (UTC)

I think you've misunderstood Zad68's point, Crimsoncorvid. The question Zad68 was addressing is "how much weight do reliable sources give HIV in the context of circumcision". This is the question that needs to be answered in order to determine whether HIV should be discussed in the lead, and if so, how much coverage should be given to it. As Zad68 points out, HIV is given considerable weight in reliable sources.

The question which you're addressing, I think, is "what is the relative prominence of the view that circumcision reduces the risk of female-to-male HIV transmission". Unfortunately, you've made several mistakes in addressing it. First, you consider which agencies have "endorsed" the HIV studies; this rests on the flawed premise that medical organisations are likely to "endorse" a study, but that is not usually the case. Second, you incorrectly state that a number of countries have rejected the studies. A few organisations have considered whether circumcision should be promoted for HIV prevention in those countries, but that is not the same issue as whether circumcision reduces the risk of f-to-m transmission. Third, you note that some have criticised the studies. While true, this isn't an argument either way, as it would be unusual to find a view with no dissenting views (even the well-established association between HIV and AIDS has its critics). Finally, you mistakenly suggest that the actual number of sources voicing a view does not count because of the influence of research funding. This is erroneous, because WP:MEDASSESS clearly indicates that the reliability of a source does not depend on its funding: "Do not reject a high-quality type of study due to [...] funding sources". Jakew (talk) 08:30, 15 June 2012 (UTC)

I see two camps of editors ... the ones supporting the status quo, the others making a thoughtful case for change. The status quo camp's argument for keeping the improbable Who studies boils down to the same argument historically made for circumcision, "It's always been that way." Debunk Da Junk (talk) 19:57, 15 June 2012 (UTC)

"Debunk Da Junk", review the Wikipedia policies regarding no personal attacks and assume good faith. Please start making comments that address the content and not the editors. Zad68 20:19, 15 June 2012 (UTC)

I sense that we're going around and around in circles. Maybe this cannot be resolved. There are studies such as this one http://www.obgynnews.com/news/top-news/single-article/mandatory-circumcision-not-cost-effective-for-hiv-prevention/72782cada61b7f416a4f282dbaf4854b.html that shows that circumcision for HIV prevention is not cost effective in the US. Similar studies show that it is not cost effective for Africa either. If not cost effective, why is it important for it to be in the lead? Your WP:MEDASSESS statement is not relevant here because that refers to deciding whether or not a source is relevant for inclusion in the article. I'm referring to what determines medical consensus. On this issue, Wikipedia policies provide little guidance. The pro-circumcision viewpoint is a minority viewpoint.

I will reiterate that as circumcision is primarily done on babies that it is not relevant to mention a study on adults in the lead. Wikipedia guidelines require us to summarize the most relevant and salient points of the article. Africa HIV claims are not the most relevant. I completely disagree with the statements made above about weight and my interpretation.

Wikipedia policy of WP:LEAD states "emphasis given to material in the lead should reflect its relative importance to the subject, according to reliable sources". This is more in line with my interpretation. The Africa HIV claims are not all that important to circumcision compared to other items that are summarized in the lead -- especially since virtually all circumcisions are performed on infants and the Africa HIV research does not cover that. Of course, everything should be based on reliable sources. Crimsoncorvid (talk) 23:42, 15 June 2012 (UTC)

As you rightly note, we're going around in circles. To avoid unnecessary repetition, therefore, I'll just address your argument re cost-effectiveness. There are basically two problems with your argument. The first problem with your argument is that lack of cost effectiveness is an argument against circumcision programmes, but it's not necessarily an argument against inclusion in an encyclopaedia. We include information about a topic, in accordance with the weight given by reliable sources, regardless of whether that information is positive or negative towards the subject. For the sake of argument, suppose that nine tenths of the literature about circumcision was about HIV, and all of those papers agreed that it wasn't cost-effective. That wouldn't be an argument against mentioning HIV; rather, it would be an argument for prominently discussing the lack of cost-effectiveness regarding HIV.

The second problem is that your assertion that circumcision is not cost effective in the US or Africa can only be reached by ignoring a large proportion of the literature. African studies have generally found that it is cost-effective (see a review of five such studies in PMID 20224784). There are too few US studies to be able to say; Sansom et al (see PMID 20090910) found that it is cost-saving. The article you cite appears to be about a study considering mandatory vs elective infant circumcision, which is not quite the same question. Jakew (talk) 07:55, 16 June 2012 (UTC)

This has been argued through and through (more than once). This article is worldwide in scope; while most circumcisions may be done in infancy (or youth) for religious or cultural reasons, there are numerous places where the operation's value in preventing HIV spread is either increasing the elections of circumcision or is generating much debate over its efficacy and health benefits. To ignore this in the lead would be dumbing down the article because some folks have a POV that they feel threatened by its inclusion. With the logic above, why don't we just have the lead say, circumcision is a Moslem practice and chuck the rest, since the majority of circumcisions are done by Moslems for their religious reasons. Oh, but we have an article for that Khitan, and this is for the wider perspective. doh... Carlossuarez46 (talk) 16:18, 16 June 2012 (UTC)