Talk:Acupuncture

Sources on acupuncture with positive results

Greetings! Here are couple of sources dealing with the positive research outcomes with acupuncture that I'd like to suggest to be taken into account in the article:

1. The effectiveness of acupuncture research across components of the trauma spectrum response (tsr): a systematic review of reviews: "Thirty-one studies were included in this review. The majority of included trials comparing true acupuncture and sham acupuncture showed a trend in favor of acupuncture. The combined response rate in the acupuncture group was significantly higher compared with sham acupuncture either at the early follow-up period (risk ratio [RR]: 1.19, 95% confidence interval [CI]: 1.08, 1.30) or late follow-up period (RR: 1.22, 95% CI: 1.04, 1.43). Combined data also showed acupuncture was superior to medication therapy for headache intensity (weighted mean difference: -8.54 mm, 95% CI: -15.52, -1.57), headache frequency (standard mean difference: -0.70, 95% CI: -1.38, -0.02), physical function (weighted mean difference: 4.16, 95% CI: 1.33, 6.98), and response rate (RR: 1.49, 95% CI: 1.02, 2.17). ... Needling acupuncture is superior to sham acupuncture and medication therapy in improving headache intensity, frequency, and response rate." (Sun, Y., Gan, T. J. (2008). Acupuncture for the Management of Chronic Headache: A Systematic Review. Anesth. Analg. 107: 2038-2047)
2. A randomised controlled trial of acupuncture care for persistent low back pain: cost effectiveness analysis: " A short course of traditional acupuncture for persistent non-specific low back pain in primary care confers a modest health benefit for minor extra cost to the NHS compared with usual care. Acupuncture care for low back pain seems to be cost effective in the longer term." (Ratcliffe J, Thomas KJ, MacPherson H, Brazier J. A randomised controlled trial of acupuncture care for persistent low back pain: cost-effectiveness analysis. BMJ 2006;333:626)

I suggest that these findings will be included into the article in order to have a more balanced view on the subject.

Ps. This also intersects with the Acupuncture and moxibustion as well, so if you haven't, please take a look at that as well! ;) Jayaguru-Shishya (talk) 18:18, 12 May 2014 (UTC)

Agree, these are excellent MEDRS's. There's also this (h/t Puhlaa): Evidence Map for Acupuncture, a medical guideline from the US Dept. of Veterans Affairs Health Services Research and Development Service. --Middle 8 (leave me alone • talk to me • COI?) 02:51, 13 May 2014 (UTC)

[User:Middle 8|Middle 8]] (leave me alone • talk to me • COI?), that's just great! It should be definately included in the article. Jayaguru-Shishya (talk) 15:26, 13 May 2014 (UTC)

Disagreed. We have newer sources for Acupuncture#Cost-effectiveness.

The other source is already in the article. Lee, Courtney; Crawford, Cindy; Wallerstedt, Dawn; York, Alexandra; Duncan, Alaine; Smith, Jennifer; Sprengel, Meredith; Welton, Richard; Jonas, Wayne (2012). "The effectiveness of acupuncture research across components of the trauma spectrum response (tsr): A systematic review of reviews". Systematic Reviews. 1: 46. doi:10.1186/2046-4053-1-46. PMC 3534620. PMID 23067573. {{cite journal}}: Unknown parameter |displayauthors= ignored (|display-authors= suggested) (help)CS1 maint: unflagged free DOI (link)

The Evidence Map for Acupuncture cited reviews which we currently use in the article. If there are any missing reviews we should use the reviews directly. QuackGuru (talk) 03:02, 13 May 2014 (UTC)

I agree that we should cite individual reviews, but reviews of reviews -- or medical guidelines, like the VA Evidence Map -- are also considered good MEDRS's. --Middle 8 (leave me alone • talk to me • COI?) 18:25, 15 May 2014 (UTC)

I think you have misunderstood something regarding the publication date of an article. There are currently two articles listed in Acupuncture#Cost-effectiveness, one from 2011 and the other from 2013. The one that I mentioned is from 2012. All of these articles are published by a different set of authors in rather a short period of time, and you can't make a conclusion that those authors who have conducted the latest research would be somehow automatically "the most right". It's typical for human sciences to obtain a large variety of different test results, and that's the reason why a lot of repetition is needed. The latest one available, however, does not signify that it would be somehow more reliable or a better one even. Therefore, it is still very well-grounded to include those two pieces of research in the article. Jayaguru-Shishya (talk) 15:26, 13 May 2014 (UTC)

However, the usual problem applies: what are the chances of someone studying acupuncture in this way, who is not already convinced of its efficacy and seeking to confirm it? Where will negative results get published? How many journals are looking to carry negative results for things that have no plausible mechanism of action anyway? For subjective symptoms, placebo effects, expectation effects, regression to the mean and so on are particularly strong, so do we see much weaker results of objectively measurable outcomes (yes we do) and what does this indicate about whether the intervention actually works (it weakens the conclusions of studies on subjective outcomes). You have to be really careful with reviews like the VA one. They are very often motivated by political pressure for a treatment that has failed to gain acceptance through the orthodox route of unambiguous good quality evidence. It has taken a long time to show that meridians and acupoints are almost certainly irrelevant, and most of the studies involved in these reviews will not have been able to properly blind for whether the needle is inserted or not. Guy (Help!) 09:38, 13 May 2014 (UTC)

@Guy - Given the numerous results showing little evidence for efficacy, somebody has to be publishing them, and it's not just Cochrane. You're certainly right about caveats such as blinding and expectation effects, but it's up to reviewers to account for these, right? On the VA source, IIRC it doesn't deviate hugely from other reviews, and one wouldn't expect unanimity. Author bias is possible, but that's speculative; I don't remember any red flags. Seems like yet another meta-analysis. I think it's reasonable to say that if we're going to exclude positively slanted things just because they're positive, then we're not fairly representing the current state of the literature (and I say this as someone who believes that WP:NOTTRUTH and WP:FLAT shouldn't be taken too far). For example, we saw within the last year both "pro (Wang et. al.) " and "con" (Colquhoun and Novella) editorials in a mainstream journal. (The latter rang truer, but the former didn't really contest a lot of its conclusions.)

As far as why study it at all, one answer I've seen is that it's highly safe and makes people feel better, so the large placebo effect is actually a plus when added to whatever its nonspecific effects may turn out to be. Yes, that would be really perverse logic if used when evaluating a new drug, but the double standard is apparently justified by acu's known lack of side effects and drug interactions. An example of this view is the editorial introduction to the "pro" and "con" editorials above, and I've heard some doctors, whose patients try it and like it, say similar things. Perverse but pragmatic. --Middle 8 (leave me alone • talk to me • COI?) 20:01, 15 May 2014 (UTC)

Publication bias? Yes, that might be true, or then it might be not. We are not here to conduct critical evaluation on such things though in Wikipedia. There are analytical tools however to evaluate possible publication bias, such as the funnel plot etc. It's a very interesting issue still; in an ideal case the scientific community will try to falsify the test results. For example, if opponents of some theory would like to falsify the former results, they would naturally have the incentive to re-run the test setting and publish any negative / statistically insignificant results acquired. When it comes to meta-analyses, of course, it is good for oneself to be aware what studies have been included in the meta-analysis, what have been left out, and why possibly? Anyway, that's a whole different field of research (and a lot of work and data collection!) and we are not here to do that in Wikipedia. Interesting subject though. Jayaguru-Shishya (talk) 15:26, 13 May 2014 (UTC)

Publication bias is not an issue with complementary medicine journals, this also occurs in conventional medicine. Also the suggestion that negative reviews would be buried is pure speculation but there is evidence of this with pharmaceutical companies duping the public [1]. The reality is, acupuncture, like chiropractic, is gaining increasingly researched with better designs and there seems to be a pattern emerging and that acupuncture is useful adjunct with specific diagnoses. This does not validate TCM theory, but it does suggest that needles inserted at various points seems to have a beneficial effect. DVMt (talk) 00:09, 14 May 2014 (UTC)

Guy, not only do I agree with the previous 2 comments, I would also point out that actual scientists, people who do science and whose reputations depend on a truthful explanation of science, would never be so cavalier as to assert that ANYTHING has been absolutely firmly established, ESPECIALLY the absence of phenomena based on an absence of evidence. Even Ernst, who is truly on a mission but nevertheless is a true believer in science, does not make the kind of absolute, religious fundamentalist type conclusions you make about the current scientific lack of evidence for some of the effects of acupuncture and TCM. Your appeal to attack any positive reviews for publication bias is absurd considering there is literally no profit-motive (patents) in the acupuncture profession the way there is in the pharmaceutical industry or even in the realm of surgery. Even with the current natural medicine loving culture we have in the states right now, schools of acupuncture are under pressure to reduce tuition and program length due to debt-to-income ratios. Almost nobody gets wealthy doing acupuncture, so who is driving the publication bias you complain about? Are you at the Vioxx page decrying publication bias? How bout Chantix? Cuz those are truly dangerous areas where money talks louder than science. Acupuncture is simply not such an area. Herbxue (talk) 07:32, 14 May 2014 (UTC)

Ah, a nice dose of common sense! Agreed with the comment above. Some editors are trying to substitute their own personal belief system and white-wash the whole profession or modality as pseudoscientific. With almost 21000 hits at Pubmed (http://www.ncbi.nlm.nih.gov/pubmed/?term=acupuncture) is it still really fair to say there is no scientific aspect about the study of using needles to promote positive physiological change? — Preceding unsigned comment added by DVMt (talk • contribs) 01:02, 16 May 2014 (UTC)

Proper scientists who do real scientists certainly do say that things have been firmly established. The problem here is that there are two parallel threads of research: one conducted by believers, looking to find evidence of the conditions for which acupuncture might work, and the other conducted by non-believers, looking to test the basis of the claims made for and about acupuncture. If you are determined to seek a mechanism by which it works, then you have to allow the possibility that the mechanism is placebo effects, expectation effects and the like: most publications on the mechanisms of acupuncture quietly ignore this rather important question. When you consider that the supposed meridians and acupoints are different between Chinese and Japanese acupuncture, you can't really carry on ignoring this. To quote one source:

Research on the nature of acupuncture points and meridians is often difficult to evaluate because of the diverse nature of the claims made, incomplete data provided in published studies and the variety of parameters involved in the assessment of these claims. Many of the studies purporting to have identified acupuncture points or meridians come from China; the role of publication bias in Chinese literature needs to be considered in light of the fact that no trial published in China from 1966 through 1995 found a test treatment to be ineffective. Obvious contradictions exist between current acupuncture practice and the historical record, as well as on the “correct” number of points and meridians reported by current practitioners of acupuncture. From an empirical standpoint—discrete structures such as acupuncture points and/or meridians would revolutionize the study of anatomy and physiology—no such revolution has occurred. Whatever the clinical efficacy of needling, there is, as yet, no convincing evidence to show that acupuncture points or meridians exist as discrete entities.

That seems pretty neutral to me, albeit not to the taste of acupuncturists. It's also in line with the absence of any credible evidence of anatomy associated with the purported meridians and acupoints.

We should not feel threatened by this sort of thing. There's no reason to believe that ancient wisdom is anything other than wrong in medical matters, the cases of historical remedies, diagnoses and treatments that turn out to be both valid and only recently unambiguously shown to be so, are few and far between. We worked out salycilic acid and quinine quite early in the development of modern medicine. In practice, it does not seem to matter where you stick the needles and it does not seem to matter if you actually insert them. Neither of these is at all surprising given the source of both ideas. I think the question of the non-existence of meridians and acupoints is settled, with only the believers continuing to plough that furrow. I don't think this is a particularly controversial view, other than with believers. Guy (Help!) 18:35, 16 May 2014 (UTC)

Who here is discussing about meridians and acupoints? We're talking primarily about systematic reviews of effectiveness. You're conflating theory and practice. You can get good results with a wrong theory. DVMt (talk) 19:29, 16 May 2014 (UTC)

If there are good results regardless of where you insert the needles, then any idiot could randomly insert needles and claim to provide medical care. Wait, there actually was a MythBusters episode on acupuncture which did precisely this and obtained good results. Tgeorgescu (talk) 11:34, 1 June 2014 (UTC)

If there are good results regardless of where you insert the needles, then any idiot could randomly insert needles and claim to provide medical care.

So this proves that you haven't read the sources I provided, really. If you'll take a look, you will see that the places where they insert the needles have been controlled already. I'd be surprised to find peer-reviewed studies where such a preliminary thing wouldn't be taken into consideration even.

I don't know about the university where you used to study, Theorgescu, but in mine MythBusters weren't a proper source. Your source based on MythBusters, shall I say, busted! ;) Jayaguru-Shishya (talk) 12:41, 2 June 2014 (UTC)

The argument has been made that Chinese and Japanese acupuncture have vastly different acupuncture points. It could be that nowadays that Chinese acupuncture system is pretty unified, but an argument can be made that it was unified by fiat, namely that Maoist authorities were embarassed by the diversity of Chinese acupuncture systems and have politically decided to teach only one version. In fact, it is hard to see that a medical system enshrined in folksy superstition would produce a clearly identifiable system of acupuncture points, i.e. that acupuncture points could be unequivocally identified through evidence-based medical research, instead of relying upon centuries-old writings for identifying them. E.g. a surgeon knows where to find the liver and this knowledge is objective since there are clear criteria for what is a liver and where to find it inside a body. As far as I know, there are no such clear-cut criteria for establishing acupuncture points otherwise than relying upon folklore. And we know that folklore varies vastly among different regions. Tgeorgescu (talk) 18:07, 9 June 2014 (UTC)

Interesting. This is mostly just speculation on your part but its worth addressing. Its true that the description of acupuncture point location has evolved, but the actual location of major points probably has not. We still use the cun measurement system, which uses divisions of space between anatomical landmarks. What has changed is that the classical texts did not use modern anatomical terminology to describe the points, while modern textbooks do. This is for the sake of accurate and efficient teaching, not "political fiat". There are not really different points in Japanese acupuncture generally, which mostly is rooted in the Nan Jing, a Han dynasty text. There are, however, multiple sub-styles of acupuncture, from various countries, that use unique points chosen with unique criteria. Herbxue (talk) 22:12, 9 June 2014 (UTC)

Original research?

This change was OR. CONCLUSION:There was little evidence that different characteristics of acupuncture or acupuncturists modified the effect of treatment on pain outcomes.[2] QuackGuru (talk) 19:54, 25 May 2014 (UTC)

It's not only OR, but actually devious. We follow the sources, and this one said "little". Good catch. -- Brangifer (talk) 20:56, 25 May 2014 (UTC)

Sourced text was replaced with original research again. QuackGuru (talk) 15:39, 27 June 2014 (UTC)

Milliongoldcoinpoint claims "aware of the edit,not OR when you quote directly from the source" but the text was not in quotation marks.. Here is the source.[3] I could not verify the claim. Did I miss something? QuackGuru (talk) 17:24, 28 June 2014 (UTC)

About nature source

I check the source from nature which define the TCM as pseudoscience. This source is [4]. Actually this article tries to refute another article which is also from nature [5]. The second one describe some opinion, one of them is to use system biology as a way to assess the usefulness of tcm. I just wonder whether it is good to use one article in nature as the view of nature journal while ignore others which are also from nature. Despite article [6] which is also from nature 448 in 2007, I see another article from journal nature [7] which describe the usefulness (for dementia) of TCM. This article was published in 2010 and stated " Sound therapeutic effects promote more scientists, domestic and abroad, to study extracts from herbal medicines. Today, a great number of compounds from herb extracts have proven to be multi-targeted, low toxicity and potent in alleviating dementia." It seems there are many articles which present different idea in nature [8]. I wonder whether to add all of these sources from nature to keep neutrally. I hope someone can check all of these articles from nature journal [9]. Now I think one editorial in nature is a neutral description in this article but whatever, add the website link for the reference is a good way for reader to follow up the source.

Thank you, you make an important point - does one article constitute the consensus view of Nature on the subject? Does Nature's view constitute the consensus of something larger (the "scientific community", the "medical profession")? Even if it did (hint, it doesn't), would that justify stating the opinion as fact? No. This is a case of cherry picking and blowing one statement way out of proportion leading to a weight problem, and a POV problem. Just state, in the body of the text, "An editorial in Nature labeled TCM pseudoscience" - problem solved. Herbxue (talk) 17:36, 27 May 2014 (UTC)

I thought the sentence needed to be modified slightly because, as you pointed out, Nature magazine hasn't completely come to the conclusion that it's pseudoscience. The article cited actually was, as you said, a rebuttal to another article in Nature that was supportive of acupuncture. Ricflairsbutt (talk) 20:50, 27 May 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 00:01, 2 June 2014 (UTC)

I am the user who start this section. My point is to neutrally indicate that“this is one editorial in Nature” because natural magazine has many articles which present different viewpoint.In the first paragraph, I added many articles from nature which have other view. For example, one of this [10] described the effective of TCM for dementia. — Preceding unsigned comment added by 72.33.63.1 (talk) 21:31, 4 June 2014 (UTC)

I noticed your post, but unfortunately the article you cited was on herbal treatment of dementia. I added an article from Nature on acupuncture which should meet your needs. Milliongoldcoinpoint (talk) 22:33, 23 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

Original research again

"Thirty-eight acupuncture trials were identified. Most studies (22/38 = 58%) found no statistically significant difference in outcomes, and most of these (13/22 = 59%) found that sham acupuncture may be as efficacious as true acupuncture, especially when superficial needling was applied to non-points."[11] The recent edit replaced source text with OR. The source does not claim 58% found that sham acupuncture may be as efficacious as true acupuncture. The "majority of" is the correct explanation. QuackGuru (talk) 20:47, 22 June 2014 (UTC)

How would you suggest it read? When you made your edit, it read "The majority of studies" which could imply that the majority of all studies everywhere state that sham equals acupuncture and that's not the case at all. Reading it again, I see the outcomes state that only 13 of the 38 studies found that sham equals acupuncture, so there is no way the word most could be used as far as I can see. Milliongoldcoinpoint (talk) 21:23, 22 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

If we do the math, 13 out of 38 actually equals 34%. If we changed 58% to 34% would that be original research, or should we just state 13 out of 38 instead?Milliongoldcoinpoint (talk) 21:26, 22 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

It is not our job to question reliable sources. "The majority of" is sourced per WP:V. It should read without the OR you added. It is better to summarise the source. I added the specific numbers to the body of the article. QuackGuru (talk) 21:47, 22 June 2014 (UTC)

It is our job to quote sources accurately and not out of context. To say "the majority of studies" without saying that it is the majority of studies within only one review, is misleading. And as a matter of fact it's not even the majority of studies within that one review. They have two sections of statistics, one group and one group within that group. Again, how should it read? I'm more than willing to compromise here, but let's not take what a source says out of context. Milliongoldcoinpoint (talk) 21:49, 22 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

"A systematic review found that the majority of "sham" acupuncture..." The text is currently not misleading.

The text is sourced but you are questioning the source again. To say "the majority of" is accurate according to the source.[12] QuackGuru (talk) 21:55, 22 June 2014 (UTC)

Well, not to quibble, but now it reads as though that one systematic review stated that the majority of all sham studies equal acupuncture. That's not the case. The majority of the studies they reviewed stated that, not the majority of all studies. Do you think it might read better if we state, "One systematic review found that the majority of the trials in the review stated sham acupuncture..." ?Milliongoldcoinpoint (talk) 22:14, 22 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

A systematic review found that the majority of "sham" acupuncture may be as efficacious as "real" acupuncture, and, therefore, the validity of traditional acupuncture theories including acupuncture point locations has been questioned.[13] The text is accurate but you are still questioning the source. There is no need to add extra words to question the source.

Other sources are summarised the same way in this article. For example: A 2011 Cochrane review found that there is insufficient evidence to determine whether acupuncture is an effective treatment for cancer pain in adults.[76] QuackGuru (talk) 22:30, 22 June 2014 (UTC)

Well, I don't want to make a bigger issue out of it than it deserves. We can leave it as you edited for now. I still hold that it's a fairly unique source in that there are two levels of data in the conclusion, and we aren't even accurately representing what the source state but rather making broad sweeping statements about sham studies in general. But I'm curious what other editors have to say. Milliongoldcoinpoint (talk) 23:59, 22 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

We should use the actual numbers, and clearly state which review those numbers come from.Herbxue (talk) 14:24, 23 June 2014 (UTC)

I have to agree with QuackGuru here, it does not say that 58% of trials found that "sham" acupuncture may be as efficacious as "real" acupuncture. However, it neither says that "the majority of sham acupuncture may be as efficacious as real acupuncture". What the source says is that 58% were statistically insignificant. Therefore, such conclusion cannot be made (for the obvious lack of statistical significance).

When studying further these statistically insignificant studies (22 studies which is the majority, while only 16 could demonstrate statistical significance), however, "sham acupuncture may be as efficacious as true acupuncture". We cannot make a conclusion though that "the majority of sham acupuncture may be as efficacious as real acupuncture". That doesn't override the statistical insignificance. Jayaguru-Shishya (talk) 15:14, 23 June 2014 (UTC)

I'd like to see the more detailed methodology, but in my understanding "most of these (13/22 = 59%) found that sham acupuncture may be as efficacious as true acupuncture" means that in those 13 studies out of 22, sham acupuncture was as efficacious as - or even more efficacious than - true acupunture [even though the difference is statistically insignificant, i.e. the result can be due to statistical randomness at the given risk level]. Therefore, I'd suggest that let's keep it A systematic review found no statistically significant difference between true acupuncture and sham acupuncture... etc. etc. Jayaguru-Shishya (talk) 15:38, 23 June 2014 (UTC)

What about if we quote all the relevant numbers? It could then read, "In a systemic review, 13 of 22 studies found that sham may be..."?Milliongoldcoinpoint (talk) 15:53, 23 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

I would prefer that - give the results clearly, thus not allowing a POV summary statement either way.Herbxue (talk) 15:57, 23 June 2014 (UTC)

Ok I changed it, and this looks perfect now. Thanks for everyone's input!Milliongoldcoinpoint (talk) 18:39, 23 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

I restored it to what it was before will this is discussed. This is not need "13 of 22 studies found that". Additionally we should be paraphrasing instead of using quotes as our text is supposed to be CC BY SA. Doc James (talk · contribs · email) (if I write on your page reply on mine) 20:37, 23 June 2014 (UTC)

We could shorten to "A systematic review found no difference between true acupuncture and sham acupuncture..." Doc James (talk · contribs · email) (if I write on your page reply on mine) 20:39, 23 June 2014 (UTC)

I agree. No difference doesn't mean that there weren't any difference in the studies at all; what it means is that even though there was difference, it wasn't statistically significant. Jayaguru-Shishya (talk) 20:51, 23 June 2014 (UTC)

Yes, the lede should be a summary. I reverted the proposal to discuss first. I'm not sure which version is better. The extreme low level details about the numbers do not belong in the lede or the body. QuackGuru (talk) 20:57, 23 June 2014 (UTC)

I reverted it back to your edit[13] which I found was good. I have no opinion about the lede or body, though. Sometimes I'm just scared that too much information is packed up to the lede, but I don't have a comment on that on this one. Jayaguru-Shishya (talk) 21:03, 23 June 2014 (UTC)

I reverted it back to your edit[14] which I found was good. I have no opinion about the lede or body, though. Sometimes I'm just scared that too much information is packed up to the lede, but I don't have a comment on that on this one. Jayaguru-Shishya (talk) 21:03, 23 June 2014 (UTC)

I would agree that it's not all that bad now, it's not completely out of context like it originally was. But the source uses the word "may" instead of just the current, definitive "no difference." I believe they might have worded it such to reflect the substantial number of studies that said otherwise. It wasn't 100-0, or even 90-10, it was 13-9. To keep all editors happy, I'm changing it to read, "A systematic review found there may be no difference between true acupuncture and sham acupuncture..."Milliongoldcoinpoint (talk) 21:30, 23 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

You added OR to the article again. QuackGuru (talk) 23:56, 23 June 2014 (UTC)

Could you please explain how it is OR when the source clearly says, "and most of these (13/22 = 59%) found that sham acupuncture may be as efficacious as true acupuncture, especially when superficial needling was applied to non-points." It says "May be" so I think not saying it is OR. I'm afraid we have to agree to disagree.Milliongoldcoinpoint (talk) 01:28, 24 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

The source does say "that sham acupuncture may be as efficacious as true acupuncture" but we are not discussing that part in the lede. We are summarising a different point in the lede.

A systematic review found that the majority of "sham" acupuncture may be as efficacious as "real" acupuncture, and, therefore, the validity of traditional acupuncture theories including acupuncture point locations has been questioned.

If you want to add "may be" to the lede this text is supported by the source. QuackGuru (talk) 01:51, 24 June 2014 (UTC)

"...sham acupuncture may be as efficacious as true acupuncture... <- and there was found no statistically significant differences in those studies (n=22). Therefore, one cannot make any conclusion from that. Jayaguru-Shishya (talk) 11:45, 24 June 2014 (UTC)

I'm not sure I'm following you. You said "may be" is in the source. Why do you object to using it in the lede?Milliongoldcoinpoint (talk) 03:34, 24 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

The part "may be" is sourced for the text the majority of "sham" acupuncture "may be" as efficacious as "real" acupuncture. The part there "may be" no difference between "true" acupuncture and "sham" acupuncture is not supported by the source. These are two separate points. QuackGuru (talk) 03:42, 24 June 2014 (UTC)

Thank you for clarifying. To me, saying "as efficacious" and "no difference" are the exact same thing. I'm curious what others think?Milliongoldcoinpoint (talk) 04:09, 24 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

Popular press peice

This is a popular press peice [15] Thus removed. Doc James (talk · contribs · email) (if I write on your page reply on mine) 22:42, 23 June 2014 (UTC)

Does not appear to be pubmed indexed or a review. Doc James (talk · contribs · email) (if I write on your page reply on mine) 22:43, 23 June 2014 (UTC)

This text [16] was supported by a pubmed indexed source [17] Doc James (talk · contribs · email) (if I write on your page reply on mine) 22:50, 23 June 2014 (UTC)

I thought it was a secondary source since the article commented on pubmed cites, but I agree the bar must be higher for this article. There are non pubmed, non review sources everywhere. I removed one above it and think I'll get to work on some more!Milliongoldcoinpoint (talk) 23:01, 23 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

Careful fellas, there's tons of these kind of sources in here. We need to be clear about what kind of text can be supported by such sources and what cannot. I would recommend identifying the text you feel is unsupported by reliable sources and discussing here. Herbxue (talk) 23:19, 23 June 2014 (UTC)

Doctor James, you removed this edit [18] was supported by these two pubmed indexed sources [19] [20] At the top of this page, it says I should assume good faith. I will also assume good faith for this edit I had to correct of yours here [21]Milliongoldcoinpoint (talk) 01:41, 24 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

Since I added a well sourced article from Nature which was backed up by two pubmed indexed sources, and it was removed, I propose adding it again. Are there any objections to its inclusion?Milliongoldcoinpoint (talk) 13:13, 24 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

Why was it removed? What did the edit summary say? -Roxy the dog (resonate) 13:55, 24 June 2014 (UTC)

Hello, the edit summary said "trimmed primary source". Doc James removed it, and you can read his reasoning above, and my reasoning below that.Milliongoldcoinpoint (talk) 14:15, 24 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

Well, the Nature article looks ok as a primary source to support clearly-attributed statements. We've been using other articles from the same source, why is it "popular press" now? Herbxue (talk) 16:54, 24 June 2014 (UTC)

Since there was support and no objections to it, and since we all agree that Nature is a worthy source, I put it into the article. Milliongoldcoinpoint (talk) 03:37, 25 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

Here is the source [22]. It is not pubmed indexed. It is a "news" article. Not a review article. Doc James (talk · contribs · email) (if I write on your page reply on mine) 03:56, 25 June 2014 (UTC)

Agree with JMH. Its a news article written by a freelance journalist (do a search on the author's name). This is very different from a peer reviewed review article written by an expert in the field. Formerly 98 (talk) 04:17, 25 June 2014 (UTC)

Well, it did meet Wikipedia's secondary source criteria for medical articles, but if the editors here feel everything needs to be peer reviewed and backed w a pmid,that's fine too, I added another nature source that has a pmid. Thanks for the input. Milliongoldcoinpoint (talk) 05:05, 25 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

Not sure why this was added or if the text is sourced. The source says: "The editor of Nature China reports on his first visit to a traditional Chinese medicine practitioner to find out how this ancient practice is dispensed in the twenty-first century — and to see if anything can be done to relieve his back pain."[23] QuackGuru (talk) 07:07, 25 June 2014 (UTC)

So far the strong claim could not be verified. QuackGuru (talk) 21:04, 25 June 2014 (UTC)

As long as we are not applying a double standard. I have been complaining about articles like the Matuk article being used to say TCM is not concerned with anatomy, or the Slate article being used to say that Mao invented TCM. Those were (wrong, and) of the same or lower level of reliability as this Nature article. Please be careful to not be cherry picking here.Herbxue (talk) 16:09, 25 June 2014 (UTC)

Mao invented TCM? LOL! What is this, somebody trying to open the gate for conspiracy theories to Wikipedia perhaps? =D Jayaguru-Shishya (talk) 16:20, 25 June 2014 (UTC)

Original research in the article again

Background information from recent misleading edits or replacing source text with OR: Milliongoldcoinpoint recently made some controversial edits. The word "some" is OR. The specific number 58 is OR. This edit added extreme low level details to the lede when the lede should be a summary. The part "may be" is OR. Sourced text was repeatedly deleted source material[24][25] but there was a compromise to use in-text attribution. There is a new account making controversial edits and there is another account that are doing very similar things to this article like adding OR[26] or low level details to the lede. Middle8 also recently made some controversial edits. Do I hear quacking?

The source says: "Serious complications after acupuncture continue to be re-ported. Many are not intrinsic to acupuncture, but caused by mal-practice of acupuncturists. This might explain why surveys of adequately trained therapists failed to yield such complications[28,147,149]. Most of the case reports originated from Asia (Tables 2–4), possibly reflecting the fact that, in Asia, acupuncture is more widely practised than elsewhere. Alternatively, it might be due to more Asian therapists being poorly trained[113]."[27]

The lede says" "Serious adverse effects (including death in rare cases[5]) are known[10], but have not been reported in surveys of adequately-trained acupuncturists.[5]"

The body says "A 2011 overview of systematic reviews (without language restrictions) found that serious complications following acupuncture have continued to be reported, but not in surveys of adequately-trained acupuncturists."

There are problems with the changes. For example, the part "but have not been reported in surveys of adequately-trained acupuncturists.[5]" and "but not in surveys of adequately-trained acupuncturists." seems like original research.

Was sourced text replaced with OR? We are going to have to review these recent edits. For example, see this diff. QuackGuru (talk) 16:30, 24 June 2014 (UTC)

Just use the actual friggin numbers from the source! Why are we still discussing this? Why are you edit warring AGAIN? Over such a minor issue. Sheesh. Herbxue (talk) 19:51, 24 June 2014 (UTC)

* QuackGuru, please do have the respect to explain why some specific edit(s) is/are OR in your opinion. Otherwise, we can't really be of any assistance.

* I find Middle8's edits to improve this article. No complaints about those IMHO. Jayaguru-Shishya (talk) 20:31, 24 June 2014 (UTC)

@ QG, you said that my wording "seems like original research", and you ask if sourced text was used. Have you compared my edits with the source? I used the very same wording, with only minor changes in grammar (for the sake of good prose) that I do not believe affect the meaning. Ernst says that serious complications (serious adverse effects) have not been reported in surveys of adequately-trained acupuncturists. That's what the article said too, both before and after my edits, which were meant to clarify and highlight this important aspect of safety. I also followed Xu's wording. No need to ask whether or not I might have, just look it up. And just what do you mean by "the duck test"? Are you accusing me of sock- or meat-puppetry with this article? --Middle 8 (leave me alone • talk to me • COI?) 06:38, 25 June 2014 (UTC)

Ersnt said "Many are not intrinsic to acupuncture, but caused by mal-practice of acupuncturists. This might explain why surveys of adequately trained therapists failed to yield such complications[28,147,149]. I think you did add OR/misleading text to the article and low level details to the lede. There has been too many recent controversial edits by more than one account on this article. I didn't notice a serious problem with the previous text. QuackGuru (talk) 06:59, 25 June 2014 (UTC)

That doesn't support your OR claim in any way (let alone your implicit sock accusation). Ernst's quote states as fact that "surveys of adequately trained therapists failed to yield such complications". Also he says that this fact may or may not explain his observation about malpractice, but whether it does or not, it's still a statement of fact in a MEDRS, and there's no OR. --Middle 8 (leave me alone • talk to me • COI?) 07:32, 25 June 2014 (UTC)

"Serious adverse effects (including death in rare cases) are known..." The wording can be improved. Saying Serious adverse effects are known does not tell the reader anything. Putting text in parentheses (including death in rare cases) is weird in the lede.

When Ersnt said it might explain... it that does not mean it is now fact. The low level misleading details do not belong in the lede. Ersnt never said but the surveys...etc. QuackGuru (talk) 07:50, 25 June 2014 (UTC)

Again, "might explain" part doesn't affect the truth-value of "surveys of adequately trained therapists failed to yield such complications". The latter is a statement of fact and not misleading. Please move on. As for "Low level detail", that's something you often bring up but it isn't in WP:PAG; please support your argument with something that is. --Middle 8 (leave me alone • talk to me • COI?) 08:25, 25 June 2014 (UTC)

I still can't find any OR in Middle8's edits, and I have to disagree with QuackGuru here: I think the edits helped to improve the article. Jayaguru-Shishya (talk) 10:21, 25 June 2014 (UTC)

There was an objection to the misplaced text but it was restored against consensus. QuackGuru (talk) 20:51, 25 June 2014 (UTC)

Surveys are not a great source of evidence. Thus we should separate these points into two sentence. Not notable enough for the lead. Doc James (talk · contribs · email) (if I write on your page reply on mine) 21:00, 25 June 2014 (UTC)

Oops, yep I realized this after looking at the sources Ernst cites for his statement (papers by himself a/o White). When he said "surveys" he was talking about surveys of practitioners as opposed to literature surveys of case reports involving adequately-trained acu'ists. thanks for fixing. .... oops, I see there's been an edit war, anyway am going to go back to the correct version.[28] --Middle 8 (leave me alone • talk to me • COI?) 04:42, 26 June 2014 (UTC)

It looks like editors followed Middle8's change to the lede (without any rationale explanation).[29][30][31][32] Middle8 previously told me to Please move on. This comment was deleted. The part "You can't cherry pick only the parts of a source that you like." makes absolutely no sense. QuackGuru (talk) 04:58, 26 June 2014 (UTC)

Time to move on QG. Your objection was vague and unclear, and you were IDHT-ing which is why I said move on. Doc James by contrast identified the problem and was clear and easy to understand. Herbxue has a right to remove his own comment, so why are you bringing it up?

Roxy, please comment on content not the contributor.[33][34]. That said I am now going to IAR and comment on contributors: it's sad the way editors line up over perceived ideologies, especially in the "borderline" areas like acu and chiro that are mixed bags. Whether editors support or oppose an edit, it's not a good idea to take a position reflexively without reading the source. I can tell that Doc James both read and understood the source I used but it's not clear who else did, so, just some food for thought. All that was needed was for somebody to say "can someone post an excerpt of the original" and we probably would have figured it out without an edit war. /grumbling --Middle 8 (leave me alone • talk to me • COI?) 06:15, 26 June 2014 (UTC)

Supporting text

This was added "However, another editorial in Nature found the actual practice of TCM to be modernized, well accepted, and scientific". A few issues

1. Where does it state it is an editorial?
2. What quotes from teh article supports the text in question? Doc James (talk · contribs · email) (if I write on your page reply on mine) 21:20, 25 June 2014 (UTC)

I could not verify the strong claim. QuackGuru (talk) 04:41, 26 June 2014 (UTC)

I have full text access to Nature and many other publications where I work. And I posted it after having read Wikipedia's policy on journal sources that require paid access; namely, that these articles are allowed, but full or partial text is preferred. I could post what the editorial says verbatim, but would I be correct in saying that would violate copyrights?

But for all you naysayers, I will post this critique from sciencebasedmedicine, who criticized Nature and the very editorial I cited, as shilling for tcm I won't be back on wikipedia until the weekend, so we can hash out the details then. Milliongoldcoinpoint (talk) 14:18, 26 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

I got a copy of the editorial, and it's report of a visit to a TCM clinic for back pain. It might be usable to describe what such a visit is like. I don't agree (at all) that it supports the sentence "However, another editorial in Nature found the actual practice of TCM to be modernized, well accepted, and scientific". (aside: I also don't see how it rises to the level of "atrocities against skepticism and science" that Gorski calls it, but then blogs don't get as many readers without a certain level of drama.) --Middle 8 (leave me alone • talk to me • COI?) 08:49, 27 June 2014 (UTC)

Is everyone sure they are looking at the correct article?? Felix Cheung is the editor of Nature China, and he wrote two separate pieces on TCM, one on getting his experience back pain treated and another on TCM's modernization and scientific merits. The two different PMID's are: 22190085 and 22190091 and the names of the articles are "TCM: Made in China" and "Modern TCM: Enter the Clinic", respectively. Milliongoldcoinpoint (talk) 15:16, 27 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

You are not providing verification for the text. I think it should be removed now. QuackGuru (talk) 15:39, 27 June 2014 (UTC)

I will remove it, but not because it doesn't support the statement made. I do not wish to set the precedent of materials posted which require full text access to verify the claim. Milliongoldcoinpoint (talk) 16:34, 27 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

Safety and weight

I've tweaked the wording again re safety in the lede and omitted death from the lede since there are a couple per year worldwide ascording to our sources.[35] Awhile back we reached a consensus that this was too little to mention and undue weight. If editors want to put it back let's figure out a way to make clear what the scale is. Also annotated a review below that found 26 cases (including 14 deaths) of cardiac tamponade: the text neglected to mention that the review was done without language or time limits and thus the sample size is all human-occupied space and time since medical reporting began.[36] We need to be objective; whatever else is going on with acupuncture it's not that dangerous. --Middle 8 (leave me alone • talk to me • COI?) 06:33, 26 June 2014 (UTC)

I disagree with your the recent changes. You claim you have consensus for your changes where there is none. What you added to the lede is a problem. For example, "(especially in developed countries[5]) and often are due to malpractice[1][5]. is low level details again. Not sure if the text is accurate and sourced. The source says there are "many serious adverse events from developed countries, including Australia, Austria, Canada, Croatia, France, Germany, Holland, Ireland, New Zealand, Spain, Sweden, Switzer-land, the United Kingdom, and the United States."[37] QuackGuru (talk) 06:47, 26 June 2014 (UTC)

Many yes, but not most -- well, it's closer than I thought, given the developed countries in Asia; I'm removing that part.[38] Now it says "Serious adverse effects are rare[1][10] and often are due to malpractice[1][5]." Consensus -- I'm not claiming any present consensus, just saying that there was an earlier time when we decided not to have death in the lede, and it was because the rate is that low: something some of our current editors may not have realized. I know a lot of things cause more than a couple of deaths per year due to people screwing them up and we don't put that in the lede. --Middle 8 (leave me alone • talk to me • COI?) 11:53, 26 June 2014 (UTC)

If you don't like the way this article is presented, please do what LeadSongDog says here—Khabboos (talk) 13:38, 26 June 2014 (UTC)

@Khabbos - Easier way: Edit article, discuss on talk page. --Middle 8 (leave me alone • talk to me • COI?) 03:40, 27 June 2014 (UTC)

I don't understand the objection to "low level details" - the facts in the sources are the only thing we can agree on. At this point I would rather see a flood of raw data (sl. exaggeration) rather than the POV summaries the QG is always cooking up. Its the only way to keep NPOV in alt med subjects.Herbxue (talk) 15:54, 26 June 2014 (UTC)

The text failed verification. "Many [SAE's] are not intrinsic to acupuncture, but caused by malpractice of acupuncturists." does not verify the claim "often". <ref name="Adams 2011"/><ref name="Ernst 2011"/><ref>{{cite pmid|24554789}}</ref> All these three refs at the end of the sentence do not verify the claim. The part "due to malpractice" does not summarise the body well. QuackGuru (talk) 06:10, 27 June 2014 (UTC)

I think it does, when you also look at tables 2-4 in Ernst '11 (ref.5). All cases of infection are preventable with CNT, and all cases of pneumothorax and cardiac tamponade are preventable with proper needling. 34 cases of infection, 21 cases of pneumothorax and 2 cases of cardiac tamponade (without comorbid pneumothorax) = 57 cases of 95 reported. Which I think qualifies as "often". Adams (ref.1) says "Many of the serious AEs might have been caused by substandard practice" but I agree that's too weak and am removing it. I don't know why another editor added Park '14, the pregnancy review; I'm removing it also pending discussion. --Middle 8 (leave me alone • talk to me • COI?) 08:38, 27 June 2014 (UTC)

I revised your most recent edits, and I have to say Good work Middle8! Jayaguru-Shishya (talk) 10:27, 27 June 2014 (UTC)

You are drawing your own conclusions when the source does not specifically say it was "often". QuackGuru (talk) 15:39, 27 June 2014 (UTC)

We're talking about how many SAE's are due to malpractice. Not only is 57 out of 95 "often", it's "more often than not", which is what I'm going to change the text to in a moment. Paraphrasing and using common sense are not only allowed, but encouraged. Remember that one of WP:5P is IAR, which is basically another way of saying "use common sense". I'm not saying that we need to invoke IAR to justify this edit, because we don't. What I am saying is that this edit is simple common sense, and indeed a very conservative one given IAR. I encourage you to think about IAR and "writing for the enemy" (WP:OPPONENT) and take them to heart. --Middle 8 (leave me alone • talk to me • COI?) 05:51, 28 June 2014 (UTC)

I could not verify the claim "and more often than not are due to malpractice, and therefore preventable with proper training."

I think the text is misleading. QuackGuru (talk) 09:33, 28 June 2014 (UTC)

Explained above and in ES --Middle 8 (leave me alone • talk to me • COI?) 11:05, 28 June 2014 (UTC)

You added original research, plain and simple. QuackGuru (talk) 11:10, 28 June 2014 (UTC)

Well, I've explained above why I think it's a reasonable translation (57 out of 95). Again, I think your insistence on the article using the precise and literal wording of the source is rigid and goes against common sense.

I've explained my reasoning above, and all you've said is "I could not verify the claim" and "you added OR", which is simply a repetition of your earlier comment "you are drawing your own conclusions". You're not addressing my arguments, but just repeating your position. This is textbook IDHT, not at all helpful in WP:DR. --Middle 8 (leave me alone • talk to me • COI?) 11:30, 28 June 2014 (UTC)

@ Jmh649/Doc James - Re your reversion of a couple of edits -- I'll briefly explain them both and then let you reply. This I explained here, referring to the tables in Ernst '11. He says "many" AE's are due to malpractice, and that works out to be at least 57 of the 95 total that he found, (searching internationally) from 2000-2009, which sounds like "more often than not" to me. .... And then this I explained in the ES itself; it obviously follows from the definition of malpractice, and the source specifically makes this connection too (cf. quote in ES). Thanks! Have a great weekend. --Middle 8 (leave me alone • talk to me • COI?) 12:11, 28 June 2014 (UTC)

So which text from the paper supports this? Doc James (talk · contribs · email) (if I write on your page reply on mine) 12:17, 28 June 2014 (UTC)

(1) "Many are not intrinsic to acupuncture, but caused by malpractice of acupuncturists" and (2) tables 2-4, from which we learn further that at least 57 of the 95 SAE's were due to malpractice (infection or pneumothorax or cardiac tamponade). Or does WP's dictum "use common sense" preclude doing arithmetic when summarizing sources?  :-) --Middle 8 (leave me alone • talk to me • COI?) 05:09, 29 June 2014 (UTC)

The source does not specifically say it was "often" or say "more often than not". The source says: "Many are not intrinsic to acupuncture, but caused by malpractice of acupuncturists.[39] I changed it from "often" to "many" according to WP:V policy.

The source does not specifically say "and therefore preventable with proper training." The source says: "In order to minimize the risk, all acupuncturists should be trained adequately."[40] I added only sourced text to the lede rather than original research. QuackGuru (talk) 03:33, 29 June 2014 (UTC)

5 deaths in 10 years is undue weight in the lede; see my comment above (diff). Additionally, Ernst '11 doesn't say "deaths continue to be reported". --Middle 8 (leave me alone • talk to me • COI?) 05:20, 29 June 2014 (UTC)

"Serious adverse events, including deaths, continue to be reported."[41] The text is sourced and is a summary of the body. QuackGuru (talk) 05:24, 29 June 2014 (UTC)

I checked and it doesn't say this anywhere in the paper. It is in the abstract online, but not the abstract as displayed in the paper. Anyway, we're failing WEIGHT if we don't make clear how frequent (or infrequent) such deaths are. 5 in 10 years. What a POV push. --Middle 8 (leave me alone • talk to me • COI?) 06:36, 29 June 2014 (UTC)

Great, since you like to stick close to the source, let's do it,[42]. ES was truncated, should say "if we're going to insist on scary TEH DEATHS in the lede, then let's be clear about scale". Having it in the lede at all is undue weight; we'd never do this in another medical article (or any other article) with that rate of deaths. Excellent example of POV-pushing from the anti-CAM side. Plenty to criticize in acu, but it's pretty safe. Too much attention is given to scary death in both body and lede; another e.g. is 2nd para of Acupuncture#English-language, which goes into all that detail for a total of 17 deaths found worldwide EVER. --Middle 8 (leave me alone • talk to me • COI?) 05:35, 29 June 2014 (UTC)

Sourced text was replaced with wording not found in the source. QuackGuru (talk) 05:32, 29 June 2014 (UTC)

Explained right above[43]. What WP policy says we have to use exact wording from source? None at all. Quite the contrary; we're supposed to use common sense and write for an encylopedia, not a link farm. You're not following WP:PAG. --Middle 8 (leave me alone • talk to me • COI?) 05:38, 29 June 2014 (UTC)

Then again, you also once said that rendering "0.05 in 10,000" as "5 in one million" was "too confusing".[44] That was... remarkable. --Middle 8 (leave me alone • talk to me • COI?) 09:55, 29 June 2014 (UTC)

Bit more on ref "Ernst 2011"[45] (I have the full paper): For 2000-2009, searching multiple databases internationally and without language restrictions, they found "Ninety-five cases of severe adverse effects including 5 fatalities". So 5 fatalities over 10 years, which I rendered as "an average of one death every two years was reported internationally." --Middle 8 (leave me alone • talk to me • COI?) 09:55, 29 June 2014 (UTC)

The source cited actually says that serious adverse effects, including deaths, continue to be reported. You removed that text and substituted your own analysis of the review. Just because the time scale for a review covers a period of 10 years, it doesn't mean that every report within that period was included. In fact we know from the abstract that inclusion criteria were employed to filter the sources considered. Within that selected group there were 5 reported fatalities, but we have no way of knowing what time period the selected group covered. In short, you're making the assumption that the fatalities considered by Ernst were the only ones reported; that those 5 were spread over a time period that you're assuming was ten years (by guessing the end date for the selection criteria); and then you've applied an average using an uncertain numerator over an uncertain denominator. Of course, it suits your POV to say "one death every two years internationally" than "deaths continue to be reported" (which is what the source says), but using unjustifiable assumptions to attempt to whitewash the broad conclusions of the source isn't how we should be dealing with topics on an encyclopedia. --RexxS (talk) 13:11, 29 June 2014 (UTC)

@ RexxS, I'm not sure you've read the review itself, because you're quoting a passage that's in the PubMed abstract only and not in the paper. (The version of the abstract in the PDF doesn't have the same wording as in PubMed.) Am I wrong? If we're editing at this granular level I'd expect editors to have the full source at hand.

I meant "reported" as in "reported by the review that did the search". Just saying "continue to be reported" without a number imparts little information (we're not obliged to include every statement from a source). I understand what you're saying, but I don't think the results in the review are very far off what was reported globally in the literature, because the goal was to find all reported deaths, and Ernst's selection criteria were as inclusive as possible. There wasn't any "filtering out" going on. (Injecting drugs into acupoints is not acupuncture.)

Ernst's selection criteria for lit search of adverse events incl. deaths

“ Embase, Medline, CINAHL, AMED, and the Cochrane Library were searched (October 2009) without language restrictions. Departmental files and bibliographies were hand searched. We included all case reports and case series relating to serious adverse effects, that is, an unwanted clinical outcome that needed medical/surgical attention or led to death. Reports of adverse effects due to injecting drugs into acupuncture points were excluded. ”

Just saying "deaths happen" without giving some idea of what we know about the scale is absurd. It certainly fits the POV of acu-critics. You assume I'm an acu-proponent, but that's not really compatible with sticking malpractice into the lede, is it, or add Cochrane reviews, or pruning inappropriate claims? I'm about as critical an acu'ist as you'll find, and expect you and other editors to follow NPA and comment on content, not the contributor, unless there's a good reason to call out an apparent POV-push. Trying to depict scale of reported deaths isn't a POV-push; just saying "death" in the lede, when the known rate is so low, is. How would you depict the rate, or would you even try? I'm going to make an edit using the exact wording from the review. --Middle 8 (leave me alone • talk to me • COI?) 13:55, 29 June 2014 (UTC)

@RexxS P.S. - OK, no need to answer; I realize you didn't read it based on your comment that I assumed what the time span was. It's not in the abstract but is in the paper (which I previously mentioned I had a copy of). If all you have is the abstract, it's a good idea to note this in talk page discussions, so that others know what information you're working with. (Especially if you're going to criticize another editor for supposedly misreading a source) :-) --Middle 8 (leave me alone • talk to me • COI?) 15:56, 29 June 2014 (UTC)

The specific details are in the body and the lede explains "Serious adverse effects are rare...." So the reader knowns as for serious adverse effects it is indeed rare. The lede should be a summary without going into excessive details. See "Between 2000 and 2009, a total of 95 serious adverse events, including 5 deaths, were reported.[5]" The context and a more detailed explanation is the body. QuackGuru (talk) 14:22, 29 June 2014 (UTC)

"How rare?" is a valid question given how rare it really is, and I repeat that with such a low number it's misleading to mention it without a qualifier, and it ideally shouldn't be there at all for WEIGHT reasons. (Because we don't mention it at all in other articles where there are more deaths.) We could go to RSN or some sort of NPOV-related noticeboard I guess. In the meantime, please consider using the wording after your first edit (partly removing the detail but leaving the number) as a compromise.[46] --Middle 8 (leave me alone • talk to me • COI?) 15:04, 29 June 2014 (UTC)

In my opinion, five death cases is far too marginal to be included in the article. I agree with RexxS however, that we cannot make our own analysis about the average annual death rate. But in the light of the source, we know that 5 deaths have been reported, and that's something I wouldn't give too much weight. In my opinion, it's too minor to be included in the article. Jayaguru-Shishya (talk) 14:48, 29 June 2014 (UTC)

"Deaths continue to be reported", and what's the information value of this? I think it's way too general and should be removed. Jayaguru-Shishya (talk) 15:11, 29 June 2014 (UTC)

@Middle 8: The source that is cited in the article supporting its content is https://www.ncbi.nlm.nih.gov/pubmed/21440191 - the abstract available at PMID: 21440191. If you want to claim that a different source contradicts it, then it's normally polite to give a clear quote for material that's not freely available. I don't usually go to the trouble of getting full text versions when the abstract is so clear, but I can if it's essential. In this case, everyone can see in the cited source where it says "Serious adverse events, including deaths, continue to be reported." What's the quote that's supporting "5 in a 10-year period"? For what it's worth, as the full text contains more detail than the summary, there's obviously going to be a case for reporting that detail in the article; but when the source's summary fails to mention a particular detail, isn't that a clear pointer against putting it into our summary?

@Jayaguru-Shishya: I disagree that fatalities resulting from treatment are "far too marginal to be included in the article". The average reader would not anticipate that acupuncture could kill people, so noting that it can happen is certainly a key point. If you can't see the information value of reporting that deaths continue from what many would expect to be a relatively harmless treatment, then I doubt that you ought to be editing in this area. Isn't it the simple truth that you're trying your hardest to whitewash acupuncture and are working to remove anything critical? --RexxS (talk) 19:24, 29 June 2014 (UTC)

Hi RexxS. No, I am not trying to whitewash acupuncture. I neither find it really nice that you are saying such as the very first thing after I told my opinion. Anyway, I was thinking that five reported death cases are not significant enough to be mentioned; we are still in the very marginal here. Jayaguru-Shishya (talk) 19:59, 29 June 2014 (UTC)

RexxS I don't think that omitting something important from the source because it is not included in the summary makes sense. It is important to include the number of deaths so the reader has a sense of scale. It would even be helpful to note death rates of other minor surgical procedures for comparison. In any case, I prefer we include the stats because numbers are less likely to advance a POV.Herbxue (talk) 20:09, 29 June 2014 (UTC)

Jayaguru-Shishya, just how many people have to die from a treatment before you think it is significant? I'm not interested in playing games with you. I look at your contributions and I have to say it does nothing to change my assessment of the nature of your edits on this article. Your most recent edits to this article include stunts like rewriting the source acupuncture may be useful as an adjunct treatment or an acceptable alternative or be included in a comprehensive management program as suggesting acupuncture as both a useful adjunct or alternative to a comprehensive management program. How can you seriously tell me that isn't a serious misrepresentation whose only purpose is to cast acupuncture in a more positive light than the source provided? --RexxS (talk) 22:58, 29 June 2014 (UTC)

@RexxS: Just to clarify, the citation is indeed for the full paper. For some reason it's paywalled when you click through via the PubMed abstract, but -- as I just discovered -- you can access the whole thing via the doi. Anyway, I guess you missed it, but I did paste in a quote from the full paper above [47] when I thought it was unaccessible. To answer your question, no, there is no good reason to assume that if a detail is omitted in an abstract, it's unlikely to be encyclopedic. If that were true then Wikipedians wouldn't care about free access journals, right? Doesn't make much sense at all, and I don't think it was a good reason for your reverting [48]; and if we're to have it in the lede, then doing it the way jps just did [49] is reasonable.

But getting back to the bigger picture and your (unduly rude) reply to Jayaguru-Shishya, it doesn't make sense to put it in the lede when it's so rare, and not intrinsic to acupuncture. All five deaths in that paper are due to some acupuncturist, somewhere, being an idiot and either using a dirty needle or sticking it in a grossly inappropriate place, like the lung. People die all the time from unsterile needles, but we don't talk about such deaths in the lede sections of (e.g.) articles on IV antibiotics. The reason for that is that such deaths aren't intrinsic to the treatment. Same logic should apply here. And the language needs tightening.... Actually, this is mostly a big improvement, except for the undue weight that comes with mentioning death in the lede. --Middle 8 (leave me alone • talk to me • COI?) 21:22, 29 June 2014 (UTC)

Middle 8, I didn't miss your quote and the quote doesn't say that "there were 5 deaths internationally over 10 years" - the fragment you quoted said simply "Ninety-five cases of severe adverse effects including 5 fatalities". I can only assume you didn't understand the point I was making that not all reports were guaranteed to be included in the review and we don't know how the five fatalities were distributed, so creating an average may well be misleading. It would be safe to conclude that there were at least 5 fatalities in a period of at most 10 years, but that's not particularly helpful to the reader.

Next, don't put words into my mouth. The mistake here is your assertion that I was asking whether a detail is encyclopedic. You need to understand that a detail may be worthy of inclusion in a Wikipedia article but not be of sufficient weight to be included in its summary (the lead). The figures that you have been battling to have in the lead may well be worthy of examination in the section on safety, but when the original authors in their summary merely tell us how many "cases of severe adverse effects including ... fatalities" were included in the review, then how can we justify extrapolating that to an average or a mortality rate "internationally" when we don't know how many events were not included in the review, or the actual time period they cover?

As for the comparison with dirty IV needles, if people have died as a result of such malpractice, and a good quality source surveyed the literature and drawn a conclusion, I'd have no problem in reporting that conclusion in the relevant article. My only concern is that we accurately summarise what the best sources say, and not make assumptions like "All five deaths in that paper are due to some acupuncturist, somewhere, being an idiot and either using a dirty needle or sticking it in a grossly inappropriate place." The fact is neither of us know how the fatalities came about and the moment we substitute our own analysis - no matter how likely - we contravene a fundamental policy and distort what the sources say. --RexxS (talk) 22:58, 29 June 2014 (UTC)

Hi RexxS, the cause of death of the five subjects is in the paper. See my comment below to Kww. More later... (I sense the heat is a little high here so let's take it down a bit OK?) --Middle 8 (leave me alone • talk to me • COI?) 10:18, 1 July 2014 (UTC)

Yes, I can see that four of the five deaths were diagnosed as pneumothorax, the other is presumably the Chang report which diagnosed "aorta-duodenal fistula, shock". It's worth remembering that cause of death is not the same as diagnosis. I still don't see how you managed to relate deaths to dirty or unsterile needles - we should be trying hard to stick to what the sources say, not what our own opinions or experiences suggest. Anyway, my point remains: none of us would expect a patient receiving treatment (of any sort) for shoulder stiffness or lower back pain to die as a direct result of the treatment. Yes, it's incredibly rare in acupuncture, but it's such a serious outcome from treating such a minor ailment that it is definitely noteworthy. Does that explain my concerns any better? --RexxS (talk) 13:54, 1 July 2014 (UTC)

Hi RexxS, thanks, I get what you're saying re: death being an unexpected and serious outcome. I agree with this; the only issue is that of proportion, relative to other topics, and how we handle those on WP. Kww said it well below. Almost anything can kill at some point just due to human fallibility and the sheer probability that goes with having 6 billion people around. Aspirin kills hundreds of people per year in the US alone;[50] undoubtedly, many of these come from misuse, which is also true of acupuncture and much else that unexpectedly causes death. IOW: one has every right NOT to expect death from generally-seen-as-benign things if one and/or one's practitioner is reasonably careful. In the US at least, acu'ists are trained to use clean needles, not to needle deeply or at all in certain areas, and so on.

Of much greater interest than raw totals is death rate. Back of the envelope (not a proposed edit; just for perspective): For acu, SAE's may be on the order of one in 200,000 treatments (per Adams, citing White), and if 5% of those are fatal (our Ernst review), then we have one in four million, ballpark. Some perspective: "A one-in-a-million chance of death can also be thought of as the consequences of a form of (imaginary) Russian roulette in which 20 coins are thrown in the air: if they all come down heads, then the subject is executed (the chance of this happening is 1 in 220, which is roughly equal to one in a million)."[51] Also see: What really has a 1 in a million chance?. See what I mean about undue weight to death in the lede?

About dirty needles: what meant to say above is that "most SAE's, including fatal ones, as in this review" are either due to dirty needles OR poor needle placement. It was the latter that that caused the 4 pneumothorax deaths and maybe the other one. (Virtually all pneumothorax from acu is preventable... just don't needle deeply... take my word for now; can source later. This is a big part of training in safety.) As for infections, most aren't fatal, but Hep C eventually could be. BTW, in the review, pneumothorax (etc.) is meant as the diagnosis of the adverse event (fatal or not); see "causality" for whether they're sure it was due to acu.

Anyway, I'm sure we can work out some compromise, and as is often the case in contentious CAM areas, it may help to get fresh eyes at (e.g.) a noticeboard. Opinions tend to get pretty entrenched among active editors in CAM areas. I think it would really help to get the article to a certain point (maybe not that far off?), and then bring in editors who specialize in helping move articles to GA status; they'll also be good judges of stuff like how to weight safety.

Sorry so long. But I think we got off to a bad start, so maybe more is better for the moment, in terms of understanding each other. (I cannot believe I just spent an hour googling and replying here... easy to get drawn in... it's fun, but I can't spend any more than 30-60 minutes per day, max, on WP.) regards, Middle 8 (leave me alone • talk to me • COI?) 13:38, 2 July 2014 (UTC)

I'm not sure, but I think from what you say above, that you now agree that the issue of fatalities is noteworthy enough to be mentioned (albeit briefly) in the lead?

The problem that I see with the back of the envelope is that you can make the mistake of dividing 200K apples by 0.05 pears and getting 4 million. Let me explain: Ernst searched the literature for reviews of acupuncture being used to treat pain. Although that's probably the commonest application of acupuncture, it's not the sole one. Ask yourself "Were there SAEs following acupuncture not included in Ernst's review of reviews?" The wrong answers are: (1) "Yes" (2) "No". The right answer is "We don't know". Similarly, ask "Were there any fatalities amongst the half dozen reports that gave no information on outcome?" The same right/wrong answers apply. In addition Ernst makes the point that "Fifty-seven systematic reviews met our inclusions criteria ... They focused on a wide range of pain syndromes ... and most were published within the last 2 years". Even though most reports were clustered in a two-year period, the deaths occurred in 2002, 2003, 2004, 2005, and 2009. Small numbers create unintuitive distributions and averages are poor indicators in those cases (check the Poisson distribution). Now, we can estimate the death rate from fuzzy figures, but it's what we call a "swag" (scientific wild-arsed guess), but I wouldn't care to assign any confidence in such an estimate. That's why I don't feel confident saying much more that "deaths have been reported" - and qualifying by describing it as "rare", rather than trying to ascribe actual figures where we really don't have the precision.

Having taught statistics for 20-odd years, half a lifetime ago, I do have some idea of odds. For example, the chance of winning the UK National Lottery is about 1 in 14 million (⁴⁹C₆), yet most weeks somebody wins. For the person who wins, it's pretty significant - and without that, nobody would buy a lottery ticket. What I'm saying is that even rare events can be noteworthy when their consequences are huge, doubly so when they are unexpected.

I don't think you're reading the causality column correctly. Take a look at the penultimate paragraph of the review and you'll see that "certain/probable/etc." implies the confidence in whether the SAE was the cause of the diagnosis. We don't have "cause of outcome" because diagnosis =/= outcome. For example in the Karst report the diagnosis was "Injury of arteria brachialis, ischemia of hand" - which was "certain" to have been caused by the acupuncture; the outcome was "Amputation below the elbow" which I contend was not caused by acupuncture (unless they used a really, really big needle). Bottom line: we end up guessing at cause of death. Hope that gives you some more perspective on what I've been concerned about. --RexxS (talk) 20:25, 2 July 2014 (UTC)

Hi RexxS; I'll address your points in the order that you raised them.

1. I agree that unexpected and rare events with huge consequences are important, but don't agree that deaths from acu are frequent enough to warrant mention in the lede. All kinds of unusual accidents happen to people and we don't mention these in every pertinent article's lede section. Again, we're talking five known deaths in a decade from acu -- and not only from treatment for pain, cf. #2.
2. It sounds like you believe the review was limited to SAE's due to pain; if so, that's incorrect. Only the search on efficacy was limited to pain. (See column "reason for acu".) The goal was to summarize reports of SAE's for 2000-2009. (Of course they may have overlooked some reports, and some SAE's go unreported, but those caveats are true in any such review.) I think the current version summarizes the source adequately, other than the undue weight to death in the lede. So my back-of-the-envelope wasn't apples-to-pears, but setting that aside: whatever the rate is, we know that -- unlike with the lottery -- we don't hear about people dying every week from acu.
3. Re cause of death (or other adverse outcome): If acu is certain to have caused the SAE and the outcome was death, the death is ascribed to acu. (So acu was the "proximate cause" of amputation, in the example you mention.) A false positive for causality would mean one less SAE ascribed to acu. But I don't understand why you raise this issue.

Sounds like we agree to disagree about whether it's a WP:UNDUE problem to mention death in the lede, specifically "five known deaths in a decade according to a comprehensive search of the international literature", which is how the source should be summarized in the article body. Anything else? Seems like we're close to moving on. regards, Middle 8 (leave me alone • talk to me • COI?) 07:13, 3 July 2014 (UTC)

RexxS, I'm glad to find out that there are some others who also understand statistics. In scientific studies, however, we never try to make findings about the whole population. There is always some criteria for selection of data, and we base our findings on that data in use. How well the included data represents the whole population, that's another issue. Especially in systematic reviews, the publication bias may be a problem, but there is whole different set of methods to study that one. In experimental trials, the small sample size might be problematic. Anyway, when we are referring to Ernst, we are just to report the findings he made on the basis of his own study; not to speculate how well it represents the whole population.

Where I agree with RexxS however, is that we certainly should not calculate any "death rates" of our own.

Where I agree with Middle8, is that five deaths are far too mariginal to be reported. Jayaguru-Shishya (talk) 13:35, 3 July 2014 (UTC)

I've looked again at the "Reason for acupuncture" column in the SAE tables (2, 3, 4) and you're right: the search for SAEs wasn't confined to just treatment for pain. That gives me more confidence in the figure of 5 deaths, although it doesn't remove the uncertainty from those cases where the outcome was unknown. I think I'd relent on my objection to the "(5 deaths reported in 10 years)" wording. I agree that the current version summarises the review adequately, but re-iterate my contention that the fact that deaths occurred from what most people would expect to be a benign procedure is of sufficient interest to be included in our summary, particularly when the authors chose to feature it in their summary. We may have to agree to disagree on that.

Cause of death is always a difficult topic to adequately summarise in our articles. I'll give you a clear example: my wife was diagnosed with bowel cancer and underwent treatment for two years; she eventually died from pneumonia. That was the cause of death, yet I know there was a story that lay behind that that couldn't be easily summarised on a certificate. I could speculate that at least one of the deaths reported in Ernst's review might be due to a botched attempt at resuscitation possibly following a spontaneous pneumothorax - acupuncture not then being the cause, proximate or distal. That would indeed mean less than 5 deaths actually caused by acupuncture in the period. The point I'm making remains: presenting 5 fatal outcomes reported as "5 deaths caused by acupuncture" involves an assumption on our part that may not necessarily be warranted by the source. That why I prefer to refer to a "small number" or a "rare event" when summarising, rather than giving a figure that carries an unsupportable precision in its implications. Hope that helps. --RexxS (talk) 15:58, 3 July 2014 (UTC)

@RexxS: That's a terrible thing that happened to you and your wife, and I don't know how long ago it was (the impact does change with time but never leaves), but you have my sincerest sympathies. Re the principle for which you mentioned that as an example (causality), I agree with your logic; I'd just like to somehow convey an idea of scale, within an order of magnitude or two, at least (can't really go down so it'd just be plus :-) ). More on this below. On including it at all, although I do agree with your logic that unexpected = important, but as with anything else, probability matters, and again I just don't think we're reaching significance (especially with respect to other things that probably happen more often but don't either, like horrible but very surgical accidents. (Any doctors in the house to comment on this? @Jmh649:/Doc James?) With respect to consensus: Myself, Kww, Herbxue and Jayaguru-Shishya have supported disclusion in the lede, and you, QuackGuru and I would guess JPS have disagreed, and I don't know about the rest. This may be worth an RfC/A. In the meantime specifying the number a la "(5 deaths reported in 10 years)" sounds like a good compromise position, to which the current text is close. But how to do this and take into account your point about causality -- which turns out to be well-founded, as the source doesn't use such language! In the abstract just says cases of SAE's were "found", and in the body talks about complications "after acupuncture". Ah! So we could say, e.g.: "A review found ninety-five cases of serious adverse events after acupuncture (including five deaths) between 2000 and 2009". That's highly faithful to the source, and superior to what we have now, which deviates in more than one respect. You may want to make a mildly bold edit; I will refrain. Over to you -- this is helpful. --Middle 8 (leave me alone • talk to me • COI?) 09:52, 5 July 2014 (UTC)

With respect to deaths from procedures, it depends on the risk versus benefit. If you have appendicitis the small risk of death from an appendectomy is not significant with respect to the risk of death from no appendectomy. With respect to having your tonsils removed after a soar throat or two the answer is not so clear and thus the risk of death from tonsillectomy is more important (note the benefits from tonsillectomy are not great). When we get informed consent for surgical procedures the risk of death is always mentioned. Doc James (talk · contribs · email) (if I write on your page reply on mine) 17:08, 5 July 2014 (UTC)

@Jmh649: Doc James, Sure. Scale is still relevant. What we know from the source is that every two years, on average, someone in Asia needled too deeply over the lung (you are explicitly not supposed to) and caused a pneumothorax, or in one case, an aorto-duodenal fistula (no idea if that was normal needling meets a congenital anomaly, or what). 5 in a decade in the world. Lede-worthy? -Middle 8 (leave me alone • talk to me • COI?) 19:11, 8 July 2014 (UTC)

I oppose mentioning any specific numbers out of context. "95 serious adverse events" could be very bad (e.g., if there were only 150 treatments given) or could be very good, if there were millions. The lead should be something closer to a number needed to harm (e.g., one serious adverse event per million treatments) than a de-contextualized, useless number. Aspirin and paracetamol, which kill far more people, don't mention specific numbers in their leads. I don't think that this one should, either. "Extremely rare" is enough detail. WhatamIdoing (talk) 00:32, 9 July 2014 (UTC)

The source does not say deaths are "extremely rare". The source does say deaths continue to be reported. QuackGuru (talk) 00:37, 9 July 2014 (UTC)

It frightens me, but I think I tend towards the pro-acupuncture camp on this one. There's a death rate associated with nearly any procedure. Remember the famous case a few months ago where the little girl wound up brain dead after apnea surgery? Would that make it legitimate to put may cause death in an article about sleep apnea surgery? Probably not. To mention this here would require that the source had demonstrated that the death rate for the procedure was abnormal compared to the death rate for minor medical procedures.—Kww(talk) 20:16, 29 June 2014 (UTC)

Yep... most of the serious adverse events in that review were due to malpractice (e.g. infection due to dirty needles, organ puncture due to needling too deeply near the heart and lungs), including all five of said deaths -- well, 4 for sure (pneumothorax); I'm not sure about aorto-duodenal fistula, which may have been acceptable needling meets a rare anatomical anomaly. Still, most SAE's and nearly all (if not all) deaths are not instrinsic to acupuncture, but rather to people's tendency to be idiots, as Dr. House would put it. --Middle 8 (leave me alone • talk to me • COI?) 22:18, 29 June 2014 (UTC)

Heh, I don't know about "pro-acupuncture camps", but just make sure you won't include me in one of those ;) Jayaguru-Shishya (talk) 20:25, 29 June 2014 (UTC)

Good reorg of lede...

Nice edit [52] from jps. Far better prose: it badly needed tightening; there were too many instances of using the near-exact wording of multiple sources (a problem severely afflicting much of the article). And even though the lede is now 5 para's, that's fine; 4 is not a rigid limit. We need more of this approach. My only significant objection is death (famous last words eh?). See Kww just above [53] and my reply [54]. --Middle 8 (leave me alone • talk to me • COI?) 22:02, 29 June 2014 (UTC)

See WP:LEADLENGTH: "The appropriate length of the lead section depends on the total length of the article. As a general guideline—but not absolute rule—the lead should usually be no longer than four paragraphs." The lede should be 4 not 5 paragraphs. I don't see a reason for 5 paragraphs. QuackGuru (talk) 23:07, 29 June 2014 (UTC)

The reason is better prose: each para has its own topic; gluing them together is awkward. I've seen articles that worked well with six or more short paragraphs in the lede. --Middle 8 (leave me alone • talk to me • COI?) 23:42, 29 June 2014 (UTC)

Per WP:LEAD we typically go with 4 paragraphs. Putting a couple of points in one paragraph is okay IMO. Doc James (talk · contribs · email) (if I write on your page reply on mine) 16:57, 5 July 2014 (UTC)

I hope nobody minds my turning the definition list into a sub-heading, but it is a help for accessibility and the section needed a break anyway. The rewrite of the lead is an improvement in my humble opinion. Although I would prefer fewer paragraphs to comply with the Wikipedia "house style", in this case it is arguable that paragraphs covering introduction, effectiveness, safety, mechanism, and history makes a coherent framework for the lead. It's worth it if we can get a consensus for that structure.

I ought to make two minor criticisms:

1. The effectiveness paragraph is too big, because it goes into too much detail. For example, the issue of variable and inconsistent evidence is worth stating, but not examining in the lead - the section on Effectiveness is where I'd expect to see the comment about publication bias. Perhaps we can look at that paragraph in the future to see if we can pare it down a little.
2. There is a sentence in the safety paragraph that concludes ... there is a low risk of adverse effects which are fairly minor. Now, in UK-English that is grammatically correct but means "the minor adverse effects are the ones which have a low risk" (i.e. it's a defining relative clause). In US-English, the construction is deprecated because "that" is strongly preferred to "which" in defining relative clauses. In either form of English, a non-defining relative clause should be separated by a comma (and I suspect that is what was intended here). I'd suggest something like ... there is a low incidence of adverse effects, most of which are minor. might summarise the text less ambiguously.

Neither of those is a serious problem, but perhaps worth consideration. --RexxS (talk) 00:49, 30 June 2014 (UTC)

Belatedly, these are both good points IMO. --Middle 8 (leave me alone • talk to me • COI?) 07:33, 5 July 2014 (UTC)

"Continue to be reported"

Can someone explain why "serious adverse events continue to be reported" is useful at all? It's from a 2009 review that we already cite (both in the body and the lede) for the "95 SAE's including 5 deaths" statistic (found in a lit search covering 2000-2009). What additional information does it provide? (Yes, the source uses that wording, but it's tautological, and editors have discretion to choose what to include, cf. this part of WP:V.) Hence this edit of mine (later reverted):

• 01:01, 1 July 2014‎ Middle 8, ES: tighten lede wording --- "continue to be reported" (per '09 source) is redundant & self-evident, given that the VERY SAME source says "Between 2000 and 2009, 95 instances [of SAE's] were reported". I trust this makes sense logically & grammatically...

BTW, we can do without this, which obviously refers to the ES above and is an example of WP:POKING:

• 00:36, 2 July 2014‎ by QuackGuru, ES: Safety: move wikilink up (be careful of an editor who may put in his edit summary something like "tighten redundant lede wording" or something like "redundant and self-apparent wording in lede".
• (edit) and then same ES again, as a null edit here on talk? [55] Why?

Content disputes should be discussed with the other party (this is obvious, but they still had to write it down in WP:DR). "Be careful of an editor..." is, in the context of a content dispute, an inappropriate comment on the contributor. This isn't a battleground. QG, if you have a concern, just discuss it here on talk, please. --Middle 8 (leave me alone • talk to me • COI?) 08:34, 2 July 2014 (UTC) edited 13:43, 2 July 2014 (UTC)

The reason why "serious adverse events continue to be reported" is useful in the Wikipedia sense is that it's a prominent conclusion of a high-quality source - by prominent in this case, I mean that the author/editor chose to include the phrase (twice!) in the summary of the review. The fact that X SAEs occurred in a recent time frame does not automatically imply that such SAEs continue to occur, so it is not true that "SAE's occurred ..." and "SAEs continue to occur ..." make up a tautology.

Could I also please remind you that the purpose of this talk page is to improve the article. Complaints solely concerned with other editors' behaviour rarely meet that objective and are best taken up on the other editor's talk page or pursued through dispute resolution. --RexxS (talk) 19:34, 2 July 2014 (UTC)

Sure it's prominent, it just doesn't add any information. The review covers 2000-2009, and it tells us that SAE's continued to be reported through 2009. (And it's very likely that they continue to be, but since we're being so highly specific about citations, I'll note that the source doesn't claim to predict the future). So? We already quantify the source's findings, so better to "omit needless words". (BTW I'm familiar with WP:TPG and so forth.) --Middle 8 (leave me alone • talk to me • COI?) 07:39, 3 July 2014 (UTC)

Surely it adds the information that SAEs were still being reported and had not ceased? If that was so mind-bogglingly obvious, then why would the authors of the review make a point of mentioning it twice in the paper and twice in the abstract? --RexxS (talk) 15:58, 3 July 2014 (UTC)

@RexxS: - Look, I'm not going to fight consensus on this if it's different from my view, but since you ask again, I repeat, I don't agree for several reasons. I think it's biased and POINT-y on the part of the authors -- verifiability doesn't guarantee inclusion, so it's not unreasonable to question -- and as scientific writing adds virtually nothing to the statement that "N # of SAE's were reported over a 10-year period". If those reports had significantly increased or decreased, we'd expect him to say so, but otherwise, it's just taken as a given that the results were distributed more or less randomly over the decade. The statement is information-poor and reads more like a press release than a useful conclusion.

Apropos of the press-release comparison, there's also evidence of bias in the source. more...

Extended content

It suggests bias to in effect give equal weight to (a) meta-analyses covering hundreds of patients and (b) single case reports of SAE's, given the very very low # of SAE's for acupuncture relative to most other treatments. I'd be surprised to see at such an approach to data in other fields. It suggest bias to not even attempt to guess at a denominator, or at least note other attempts to do so, and come up with some guess at a rate, even if it's a very rough guess. Now, I understand where Ernst is going with the comparison: that maybe acu has zero specific effects, so there's no justification for using it as long as the rate of SAE's is nonzero. (I'm not going to be disingenuous and fail to address the best arguments that I see on either side of a debate.) But there are counterarguments. First, he gives short shrift to the positive evidence for efficacy that exists, assuming that inconsistency among reviews' conclusions implies likeliness of the null result, when that's not necessarily so; heterogeneity of study designs could account for a mix of Type 1 and Type 2 errors, each of which some reviews take into account more than others. Second, his argument that it makes no sense that acu would treat some but not all kinds of pain is silly on its face (different mechanisms for different kinds of pain, no?). Third, what Ernst and many acu-critics are neglecting is that even if acu is mostly non-specific, it's highly likely that it has local effects for the same reasons massage does (such as trigger point release). Fourth -- and I know this view is controversial, but it's essentially what e.g. Vickers and the German body that acted upon GERAC said -- as long as there's no common way in practice to leverage non-specific effects (strong placebo effects that result from the ritual of acu treatment and noxious stimulation), the marginal difference between acu's placebo effects and the next best safe thing counts for something in terms of reducing suffering. That last argument is particularly valid to the degree that acu has specific (even if local) effects, but if it's all a nonspecific dance, it gets dicey.

We have discretion as editors and are not bound to include material that's flawed. I'd be surprised if enough editors join me on this to make a consensus, and some won't read it. But those are my specific reasons for objecting to this wording, fwiw. --Middle 8 (leave me alone • talk to me • COI?) 07:12, 5 July 2014 (UTC)

Duplication

A 2012 meta-analysis conducted by the Acupuncture Trialists' Collaboration found "relatively modest" efficiency of acupuncture (in comparison to sham) for the treatment of four different types of chronic pain, and on that basis concluded it "is more than a placebo" and a reasonable referral option.[59]

A 2012 meta-analysis for chronic pain stated that significant differences exist between true and sham acupuncture, indicating that acupuncture is "more than a placebo."[161]

We should not repeat the same thing twice, especially when it is disputed. See: "Commenting on this meta-analysis, both Edzard Ernst and David Colquhoun said the results were of negligible clinical significance.[60][61]" QuackGuru (talk) 17:08, 27 June 2014 (UTC)

They are in two different sections, with different emphases. Are you saying that the article shouldn't have duplicate citations anywhere, because I see dozens we could trim if that's the case. Milliongoldcoinpoint (talk) 17:52, 27 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

A compromise was reached with the wording. IMO the disputed text should not be added to different sections. You did not include: "Both Edzard Ernst and David Colquhoun said the results were of negligible clinical significance.[60][61]" QuackGuru (talk) 18:00, 27 June 2014 (UTC)

Maybe I'm not just following here, but where is the duplication? Jayaguru-Shishya (talk) 20:58, 27 June 2014 (UTC)

The same source is being used twice in different sections for very similar statements. QuackGuru (talk) 21:01, 27 June 2014 (UTC)

So...? Can you please present it now here? Just point out which sections and what parts, it'd be enough thanks. Jayaguru-Shishya (talk) 21:11, 27 June 2014 (UTC)

I already did show what was the duplication in this thread. The text is already in the Acupuncture#Pain section but now very similar text was added to another section. QuackGuru (talk) 21:17, 27 June 2014 (UTC)

Seems reasonable. I same thing is dealt with in more depth at Acupuncture#Pain. I removed the duplicated part from the article[56]. Jayaguru-Shishya (talk) 19:26, 28 June 2014 (UTC)

No consensus for disputed source

This source that was tagged was deleted from GERAC by User:QTxVi4bEMRbrNqOorWBV[57] and others.[58][59] It was not a German review. It was a review by a Chinese source and the text is not neutrally written. I recommend we go back this version before the disputed text and original research was added to the article. QuackGuru (talk) 17:08, 27 June 2014 (UTC)

I have to say I agree with QuackGuru here; I'm seeing misleading wording inserted, original research, and poor sourcing. We should return to the version from a few days ago, as suggested. Dawn Bard (talk) 17:44, 27 June 2014 (UTC)

I checked the edit history again. The only clean version I could find is this version. QuackGuru (talk) 17:50, 27 June 2014 (UTC)

No, don't undo the work of several editors on multiple parts of the article just to undo a couple edits you find questionable. Do them one by one. You are doing this because several of us have disagreed with your recent work and you want to sneak around us by undoing everything that has been done. That's edit warring. I have no reason to believe you are doing this in good faith. Undo one questionable edit at a time like everyone else does. Herbxue (talk) 17:52, 27 June 2014 (UTC)

There has been numerous questionable edits. Continuing to replace sourced text with OR is a problem. I'm not making the controversial edits or adding original research to the lede. On the contrary, you continuing to edit against consensus. You are not helping to fix the problems.[60] See Talk:Acupuncture#Duplication. You are restoring disputed text and you were ignoring the problems. QuackGuru (talk) 18:17, 27 June 2014 (UTC)

Your diffs don't really make your point (I'm not really sure what your point is). My point is there has been much work done to the article beyond the addition of sources you are complaining about. Going back to an earlier version is destructive and unnecessary. Just challenge the problematic sources one by one.Herbxue (talk) 18:56, 27 June 2014 (UTC)

You think duplication is appropriate?[61] QuackGuru (talk) 19:04, 27 June 2014 (UTC)

What does that diff (about POV tags) have to do with duplicating citations? Are you just randomly linking to make it look like you are making cogent points?Herbxue (talk) 19:10, 27 June 2014 (UTC)

You removed the tag without removing the duplication. Part of the problem is that you don't see there is any problem with the current version even after diffs was provided. QuackGuru (talk) 19:41, 27 June 2014 (UTC)

QuackGuru, what is your point? I am sorry but I can't find any. Please do have the respect for other editors to explain what you are trying to say with those innumerable diffs you give.

So far, I found no problem with the edits, and I think good job has been made. The article has improved again a bit. Jayaguru-Shishya (talk) 19:53, 27 June 2014 (UTC)

But I have found problems with the recent edits. The OR, MEDRS violations,[62] and duplication should be removed. QuackGuru (talk) 20:18, 27 June 2014 (UTC)

Without fixing the problems the tags were removed. I did explain about the problems[63]. The two sources were not MEDRS complaint. QuackGuru (talk) 20:41, 27 June 2014 (UTC)

Thanks for your reply QuackGuru. So far, OR, MEDRS and duplication have been mentioned. Now, can you please explain us why do you think a specific edit is OR/MEDRS/duplication in your opinion? Please include the specific edit and your opinion why it falls under one of those categories. Thanks. Jayaguru-Shishya (talk) 20:46, 27 June 2014 (UTC)

The edit summary claims "Baloney? Maybe, but please do provide an explanation before inserting these "FV ... OR ... MEDRS" tags all over the article. Please explain why do you think so?". But I am discussing the problems. QuackGuru (talk) 20:51, 27 June 2014 (UTC)

Questioning of Sources

On the page, many sources to edits I made have been questioned. Yet these edits are cited by pubmed indexed, peer-reviewed publications, such as Medical Acupuncture. I have counted the number of times an Edzard Ernst review has been cited, which is 28 times individually with many duplications in the article. Is this a problem of weight? Ernst has many reviews on the CAM field, and not once has he had positive findings. Ernst has been criticized many times and published commentaries on his work have have listed many problems, including selection bias, negligence, inconsistencies and inclusion of unverified data. Has the time come to tag everything Ernst claims as well? Milliongoldcoinpoint (talk) 18:10, 27 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

For controversial or strong claims the consensus was to be leave out Medical Acupuncture. See comment on 19:36, 10 May 2014: "Does Med. Acupunct. have legit peer-review? AIM, it turns out, doesn't, disappointingly: see WT:MEDRS archived discussion. We have to use good journals, no question."[64] QuackGuru (talk) 18:29, 27 June 2014 (UTC)

OR, FV, MEDRS... a lot of tags have been inserted again over the article. I'd suggest that QuackGuru provides each specific edit, one by one, and tells what he thinks is wrong in there. Mere statements like "FV" doesn't give us any further understanding. Please communicate, explain. Jayaguru-Shishya (talk) 19:57, 27 June 2014 (UTC)

We can start here. I disagree with this edit. "FV" stands for failed verification. I did explain the problems with the recent edits in the other sections of this the talk page. QuackGuru (talk) 20:33, 27 June 2014 (UTC)

Thanks for your answer. Now, can you please explain that why do you find that certain edit as FV? E.g. "In my opinion, it fails verification because..." Jayaguru-Shishya (talk) 20:48, 27 June 2014 (UTC)

It did explain this before. It was not a German review and the source was disputed. This source was deleted from GERAC by User:QTxVi4bEMRbrNqOorWBV[65] and others.[66][67] QuackGuru (talk) 20:55, 27 June 2014 (UTC)

Instead of giving those diffs (I don't really get your point), could you just please explain? Please try to paraphrase your main points what you are trying to say. I see a lot of diffs you are giving, but I don't really know what you are trying to tell with those. Thanks. Jayaguru-Shishya (talk) 21:08, 27 June 2014 (UTC)

The diffs show[68][69][70] there was no consensus to use the Chinese source on Wikipedia. There was a long discussion about this at GERAC. The result was no consensus. QuackGuru (talk) 21:13, 27 June 2014 (UTC)

Thanks, answered my question finally. I won't say yes or no to your earlier discussion at GERAP at this point though. Jayaguru-Shishya (talk) 21:35, 27 June 2014 (UTC)

I even tried to add the source to the GERAC page but I was reverted. QuackGuru (talk) 21:37, 27 June 2014 (UTC)

QuackGuru, I find it rather interesting you chose to disagree with my clearly mistaken edit, opposed to this edit. If you have a problem with what you tagged, you can do me and everyone else the courtesy of explaining each need for a tag. You tagged each source, you should be able to explain each need for a tag. A broad generalization doesn't work when each source is unique. Nor does cherry picking one source to criticize doesn't work when you tagged many sources.Milliongoldcoinpoint (talk) 21:02, 27 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

I already explained about the Med. Acupuncture source. See WT:MEDRS archived discussion. And I already explained about the other source which is not a review. QuackGuru (talk) 21:13, 27 June 2014 (UTC)

This source is a secondary, opposed to the typical systematic review. It is not a primary source, and therefore is allowed. Some other editors might feel that Medical Acupuncture isn't peer reviewed, and they are entitled to their opinion. But facts are facts, and I invite anyone to show me unequivocal proof that this is not the case. I read the archived discussion and saw no such proof. That's only two, you tagged many more. What is your argument for those, or should their tags be conceded?Milliongoldcoinpoint (talk) 21:44, 27 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

That source is not compliant with MEDRS. You are using poor sources to argue against better sourcing. Medical Acupuncture was previously discussed.

Smith, CA; Hay, PP; Smith, Caroline A (17 March 2004). Smith, Caroline A (ed.). "Acupuncture for depression". Cochrane Database of Systematic Reviews. 2004 (3): CD004046. doi:10.1002/14651858.CD004046.pub2. PMID 15846693. Archived from the original on 21 April 2008. Retrieved 2 May 2008.

We are currently using a Cochrane review for depression. You should not continue to repeatedly add poor sources to argue against MEDRS compliant reviews. QuackGuru (talk) 21:54, 27 June 2014 (UTC)

Have trimmed this "A 2013 systematic review found high-level evidence to support the use of acupuncture for major depressive disorders in pregnant women. In a 2014 secondary analysis, patients with depression and comorbid pain recovered sooner with acupuncture than patients receiving either counselling or usual care. " The ref for the first one has a very low impact factor and the second is a primary source. Doc James (talk · contribs · email) (if I write on your page reply on mine) 22:34, 27 June 2014 (UTC)

The previous discussion resulted in no consensus for the Chinese source. See Wikipedia:Fringe_theories/Noticeboard/Archive_40#German_acupuncture_trials. QuackGuru (talk) 22:39, 27 June 2014 (UTC)

We have a much higher impact factor here [71] Doc James (talk · contribs · email) (if I write on your page reply on mine) 22:46, 27 June 2014 (UTC)

The Med. Acupture source was restored against consensus. QuackGuru (talk) 00:03, 28 June 2014 (UTC)

Hi, QuackGuru, it was not the Med Acupuncture source, it was a different source. Further, I did not give my consensus to remove Med Acupuncture as a source. It is a peer reviewed journal and it meets Wikipedia's standards. Milliongoldcoinpoint (talk) 01:44, 28 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

The source you added is from Med. Acupuncture. See (PMID 24781054). The source is not a systematic review. QuackGuru (talk) 02:05, 28 June 2014 (UTC)

Interesting edit

This edit was interesting. I'm going to go read the source again. QuackGuru (talk) 22:15, 27 June 2014 (UTC)

Sceptic source

The previous discussion resulting in keeping the source. I believe the IP who made this edit is User:Klocek User:Klocek was indef-blocked. QuackGuru (talk) 00:05, 28 June 2014 (UTC)

Another source from Nature

I could not determine if the source meets MEDRS. For such claims we would need a review. I thought this source was previously deleted. See Talk:Acupuncture#Popular press peice. QuackGuru (talk) 00:09, 28 June 2014 (UTC)

Hi QuackGuru, I updated the nature source, fixing the doi and pmid. Just thought you'd like the know. Milliongoldcoinpoint (talk) 01:38, 28 June 2014 (UTC) Sock comments stricken. -- Brangifer (talk) 06:14, 30 June 2014 (UTC)

I fixed the ref but it was a MEDRS violation. QuackGuru (talk) 01:50, 28 June 2014 (UTC)

Sham acupuncture in Germany but not acupuncture in general

The recently added text is a WP:WEIGHT violation: However, a 2013 German review of sham controlled clinical trials found problems in their conclusions and stated not enough evidence to claim "acupuncture and sham acupuncture have no difference in treatment effect" and "acupuncture is just a placebo effect."[160][72]

The text claims it was a German review. It was not a German review of sham controlled clinical trials in general. It was a Journal of Traditional Chinese Medicine review of sham acupuncture in Germany specifically. The text is not neutrally written and using this source is a WP:WEIGHT violation because it is only about the results from acupuncture in Germany specifically. Also the part "However" is original research. The previous sentence and this sentence are not directly related to each other, anyhow.

This source was previously deleted from GERAC[73][74][75] The diffs show[76][77][78] there was no consensus to use the J Tradit Chin Med. source on Wikipedia. There was a long discussion at the GERAC talk page. The result was no consensus. There was also a previous discussion that resulted in no consensus for the J Tradit Chin Med. source. See Wikipedia:Fringe theories/Noticeboard/Archive 40#German acupuncture trials. I tried to include this source before (along with other sources) but I did not gain consensus to use it. QuackGuru (talk) 17:15, 28 June 2014 (UTC)

Why WP:WEIGHT? Are you suggesting that it's presenting minority views? Even if it happened deal wit German results specifically, it does certainly not mean that it is presenting "minority views". Jayaguru-Shishya (talk) 19:20, 28 June 2014 (UTC)

It is presenting only the minor view of sham in Germany rather than the broader topic of acupuncture in general. If you want to use the source you would move the source to the GERAC page rather than include it in this article. This is the wrong article for the source. This article is not about German acupuncture. QuackGuru (talk) 20:02, 28 June 2014 (UTC)

This statement is more than five years old and is provided solely for historical purposes

This edit restored the text from this source. The outdated source from 1997 says: "This statement is more than five years old and is provided solely for historical purposes. Due to the cumulative nature of medical research, new knowledge has inevitably accumulated in this subject area in the time since the statement was initially prepared. Thus some of the material is likely to be out of date, and at worst simply wrong. For reliable, current information on this and other health topics, we recommend consulting the National Institutes of Health's MedlinePlus"[79] QuackGuru (talk) 20:15, 28 June 2014 (UTC)

Correct me if I'm wrong of course, but translating QG's red text above, the source itself admits it is outdated, as I stated in my edit summary, and has no place on our page. I look forward to Jaya-S reverting himself. -Roxy the dog (resonate) 22:19, 28 June 2014 (UTC)

True, seems reasonable. I think that piece of text is already removed from the article. Jayaguru-Shishya (talk) 15:40, 29 June 2014 (UTC)

Impact factor

I have trimmed in this edit [80] "A 2014 review found that while acupuncture did not improve the in vitro fertilization (IVF) pregnancy rate when performed only at the time of embryo transfer, there was a statistically significant benefit when performed at other, specific, times." as the journal has an impact factor of 1 [81] Doc James (talk · contribs · email) (if I write on your page reply on mine) 23:36, 28 June 2014 (UTC)

A massive revert - What just happened?

Wow, what just happened here[82]? A massive revert indeed, but what was reverted and why? =P Jayaguru-Shishya (talk) 15:33, June 29, 2014‎ (UTC)

Everything from the last point that Klocek began editing again was removed. If there was some small amount of valid material lost in the reversion, I trust that responsible editors can add it back.—Kww(talk) 15:36, 29 June 2014 (UTC)

I don't know whether that was reverted by a new user or an indefinitely blocked user, but if it is a new user, this is my advice: You have to cite references for every sentence you plan to add to any article of wikipedia and avoid removing sentences with references. This article is written as per the policies of wikipedia which is that all complementary and alternative medicine articles should be written, not from the perspective of its advocates/practitioners, but from the perspective of 'researchers and scientists'. If you want to complain about wikipedia's policies, please do what LeadSongDog mentioned on the Talk:Homeopathy page which I'm linking to here (and tell me also about it, on my discussion/talk page). I'm probably the only sympathiser you'll find here, so please follow my advice or else you will get blocked, banned or topic banned (from this article).—Khabboos (talk) 16:28, 29 June 2014 (UTC)

What was reverted was all changes made after an indefinitely blocked sockpuppet began making a large series of edits. The article was restored to a previous version. This is often done when an editor who winds up blocked makes multiple changes to an article. It is then up to involved editors to redo valid edits that were lost in the revert.

Adding a link to a largely unrelated discussion repeatedly is not needed. If your point is follow policy or try to change policy, say that. It's generally not necessary, if specific policy is not being followed point out the specific edit that doesn't follow policy, indicate the specific policy and explain how it doesn't follow that policy.

Short version, stop spamming the same comment on multiple talk pages. - - MrBill3 (talk) 17:09, 29 June 2014 (UTC)

Quack Guru proposed this massive revert but did not make a good faith effort to get consensus to do so. The questionable edits could have been removed (or better, discussed) one by one. I asked him not to do it on his talk page. Where is the consensus to do this? Since QG was banned for a bit, shall we revert to a version before QG started editing? Herbxue (talk) 20:29, June 29, 2014‎ (UTC))

This is standard practice. It is usually nearly impossible to fix these types of things one edit at a time, so a mass revert is done. Klocek's duplicity and use of many sock puppets has been exposed here: Wikipedia:Sockpuppet investigations/Klocek. See this too: Category:Suspected Wikipedia sockpuppets of Klocek. -- Brangifer (talk) 00:54, 30 June 2014 (UTC)

All the "valid material lost in the reversion" was restored. I checked all the recent and past edits and updated the article. QuackGuru (talk) 22:40, 29 June 2014 (UTC)

Even if it is standard practice, shouldn't there be consensus?Herbxue (talk) 04:28, 30 June 2014 (UTC)

It's policy: WP:BAN#Bans apply to all editing, good or bad. The reason is precisely this kind of event, where someone violates his block, damages an article nearly beyond repair, and other editors then try to claim that consensus is necessary to undo the damage.—Kww(talk) 05:10, 30 June 2014 (UTC)

Hmmm, I see one suspected (proven?) sock of klocek that was active here, who made a handful of very noticeable edits. During that time Middle8 and others made many well sourced edits. Are you telling me that all of Middle8's work was restored? If so that is fine, but it would have been WAY easier to just undo the edits Milliongoldcoin or whatever the user name was, rather than to revert everything and then try to redo all the good faith edits. And forgive me for being doubtful that QG faithfully restored other editors' work, I just don't buy it, but I don't have time to check it. That's why I am opposed to this overly bold move. Herbxue (talk) 05:55, 30 June 2014 (UTC)

You're walking a very lonely road on this one. This is how we do it, so get used to it. You are welcome to check the contribution history and see if any good content has been lost. Then try to restore it. -- Brangifer (talk) 06:02, 30 June 2014 (UTC)

Not satisfying in the least. You're saying "my questionable skeptic buddy gets to undo ANYTHING he wants and its up to you to redo all the work he messed up" - I just can't accept that. You need to get support before undoing everyone else's work, especially when you are a previously banned editor with an ownership problem. And as far as being "used to" disrespectful editing, I have been since 2010 so don't worry about that, doesn't mean I'm not gonna call you on that dismissive and unhelpful response, or QG on his underhanded tactics. Herbxue (talk) 06:20, 30 June 2014 (UTC)

I don't think following standard practice for reverting to previous version before multiple edits by indef banned editors needs to go through a consensus seeking first. This technique enjoys widespread support on WP in general to deal with this type of issue. I fully sympathize with the frustration it can cause and the tedium of finding and replacing any worthwhile edits. However it has been stated that all such edits were restored. An assertion of a problem with this instance should be accompanied by some example of something undone and not redone, otherwise what is the point? Without an example we should AGF that the assertion that all worthwhile edits have been restored is true. I understand there have been conflicts with the editor who did the revert and understand some strain to AGF but going to "questionable skeptic buddy" and "underhanded tactics" without any evidence is uncalled for. - - MrBill3 (talk) 11:28, 30 June 2014 (UTC)

Herbxue, you seem to be of the opinion that I am at QG's beck and call. I'm not. I didn't act until I saw enough evidence at the SPI to persuade me that Milliongoldcoinpoint was Klocek. Since then, a checkuser has been run that confirmed the relationship. I reverted all of his edits everywhere, as I do with every sock that I block. There was nothing underhanded here except Klocek using sockpuppets to evade his block.—Kww(talk) 17:00, 30 June 2014 (UTC)

Sorry Kww my beef is with QG and my recent comments were directed at Brangifer, not you. I was over the top, but I think he understands my frustration. Herbxue (talk) 17:20, 30 June 2014 (UTC)

Look at this sock edit. I removed this (PMID 22243605). QuackGuru (talk) 23:51, 30 June 2014 (UTC)

What's wrong with the source? I understand the sock edit has to go, any problem with the source? Herbxue (talk) 00:43, 1 July 2014 (UTC)

There may be a problem with the source: "The most recent meta-analysis appearing in Fertility and Sterility on acupuncture was reevaluated in view of the marked heterogeneity of interventions, controls, data analysis, and timing of interventions in the trials that were included. After removing some of the trials and data based on more rigorous standards for a high quality meta-analysis, a significant benefit of the intervention could no longer be shown."[83] QuackGuru (talk) 04:32, 1 July 2014 (UTC)

From what I understand about an article being NPOV, we should use the source Milliongoldcoinpoint inserted here, as well as put what you, QuackGuru mention, right?—Khabboos (talk) 15:07, 1 July 2014 (UTC)

I think you should read this about the block evasion. QuackGuru (talk) 17:27, 1 July 2014 (UTC)

Commenting about other editors

This needs to stop. There should be no comments about other editors here. This page is about discussing how to improve this article. Suggested improvements should be based on high quality sources per always. And RfC can be used when there are disagreements. Discussing issues around socks can take place on peoples talk pages or some notice board. The same applies to issues around individual editors. Doc James (talk · contribs · email) (if I write on your page reply on mine) 02:41, 4 July 2014 (UTC)

Seconded, I have been rather dismayed with the various comments that are borderline uncivil that I have seen lately. I think a lot of editors, and I'm not pointing at anyone in particular, have been putting comments out there that are far from constructive in tone. As Doc James said, it's fine to disagree, but there is no need to be snippy with one another. If you feel passionate about something and it's getting heated, do yourself (and all of us) a favor and take a walk, take a few hours away from the computer, and go cool off. An integral part of Wikipedia is etiquette and because many editors are not being mindful of this, we have been losing valuable editors. If you're not convinced that this applies to you, review some of your older comments and ask yourself if you could have phrased the comment in a nicer or more polite way. We are here to improve an encyclopedia, we are not here to defame character or to push a POV. If you're not here for these reasons, then it's time to reflect on what you're really doing here. Please, stick to talking about the content and quit arguing over who did what or whom you don't trust. Seriously, enough already. The only discussion that should be taking place on this page is one about the acupuncture article. Rant over. TylerDurden8823 (talk) 03:26, 4 July 2014 (UTC)

Thirded; I recognized my role in this when I read Doc's comment on QG's user talk. For example [84], [85], and [86] (the latter not wholly unjustified for the same reasons the preceding comments are, and the other two perhaps having some wiggle room given that engagement on user talk pages is in some cases unfruitful, meets IAR). But I like the course change proposed (including the clean slate idea) and for the sake of editing environment will minimize such comments. It's long past time we got this page to the point where we can solicit broader input and get it to GA status, and that won't happen unless we focus on what matters. --Middle 8 (leave me alone • talk to me • COI?) 07:28, 5 July 2014 (UTC)

"Stated it is X", "described it as X"

Note: this thread has been moved from Wikipedia talk:Manual of Style/Words to watch, and is about this WP:BRD series of edits:

• 01:09, 4 July 2014: (emphasis added) change "An editorial in Nature stated that TCM is largely pseudoscience, with no valid mechanism of action for the majority of its treatments" to "An editorial in Nature characterized TCM as largely pseudoscience, with no valid mechanism of action for the majority of its treatments." ES: minor wording change: "stated" ==> "characterized" -- neutral synonym, better prose (we always seem to say "stated", good to vary it
• 06:32, 4 July 2014 Revert; ES: (Reverted to revision 615313825 by QuackGuru (talk): Reverted non neutral change characterised as neutral. (TW)
• I posted at the user talk of the reverting editor (see here), and without more to go on other the knowledge that the concern was NPOV, posted at Wikipedia talk:Manual of Style/Words to watch. After some discussion (at the point of this version) I made this edit:
• 23:57, 4 July 2014‎: change "An editorial in Nature stated that TCM is largely pseudoscience, with no valid mechanism of action for the majority of its treatments" to "An editorial in Nature described TCM as largely pseudoscience, with no valid mechanism of action for the majority of its treatments."

Further discussion ensued. This latest version is pasted below. I'm manually annotating pre-existing signatures to indicate discussion was on that page. (whew) --Middle 8 (leave me alone • talk to me • COI?) 12:55, 5 July 2014 (UTC)

"Stated" is a universally-agreed-upon neutral verb, but we use it so much that prose can suffer, so neutral synonyms are desirable. When somebody states that "X is like Y", I assume it's also neutral to say they're "describing" X as Y, or "characerizing" X as Y. Does either of these seem problematic? (Context: this was reverted, which surprised me. But the topic area is highly fraught.) Thanks. --Middle 8 (leave me alone • talk to me • COI?) 10:28, 4 July 2014 (UTC) originally posted at Wikipedia talk:Manual of Style/Words to watch; moved 12:55, 5 July 2014 (UTC)

"Characterize" is not strictly a synonym for "describe". To characterize means to "define the character or identity of, to mark, distinguish; to be typical or characteristic" (OED). Thus it's a bit less direct than describe/call. Alexbrn ^{talk|contribs|COI} 10:44, 4 July 2014 (UTC) originally posted at Wikipedia talk:Manual of Style/Words to watch; moved 12:55, 5 July 2014 (UTC)

A good synonym for "characterize" would be "demarcate", which is exactly what people do when they call something pseudoscience, which is why I thought it was a decent choice. Oxforddictionaries.com (OED's cousin) defines "describe" as "Give an account in words of (someone or something), including all the relevant characteristics, qualities, or events", and defines "characterize" as "Describe the distinctive nature or features of", so they're pretty close. --Middle 8 (leave me alone • talk to me • COI?) 11:19, 4 July 2014 (UTC)) originally posted at Wikipedia talk:Manual of Style/Words to watch; moved 12:55, 5 July 2014 (UTC)

Demarcation is a different thing again: setting the dividing line between things. Alexbrn ^{talk|contribs|COI} 16:57, 4 July 2014 (UTC)) originally posted at Wikipedia talk:Manual of Style/Words to watch; moved 12:55, 5 July 2014 (UTC)

I shouldn't be saying "synonym" when I mean "cousin". "Characterize" isn't the same thing as "describe", but a "word to avoid" in this context? There's a sense of "less direct", but it's also more specific. It seems like a good choice, because it's got some.... characteristics in common with "demarcation". But I guess it seems too close to hedging and therefore not neutral. "Described" is probably better, and I think it's better than "stated".--Middle 8 (leave me alone • talk to me • COI?) 23:51, 4 July 2014 (UTC)) originally posted at Wikipedia talk:Manual of Style/Words to watch; moved 12:55, 5 July 2014 (UTC)

note - at this point I made the third edit mentioned above, 23:57, 4 July 2014‎. --Middle 8 (leave me alone • talk to me • COI?) 12:55, 5 July 2014 (UTC)

Maybe we should invite Roxy the dog to this discussion via WP:Echo so that Roxy the dog can elaborate on the reasons for reverting? Flyer22 (talk) 02:47, 5 July 2014 (UTC) originally posted at Wikipedia talk:Manual of Style/Words to watch; moved 12:55, 5 July 2014 (UTC)

@Flyer22: - The only reason I didn't is because I didn't think he was interested after this. --Middle 8 (leave me alone • talk to me • COI?) 13:16, 5 July 2014 (UTC)

This showed up on my watchlist: "re two edits down -- "described" may be best fit, cf. MOS discussion Special:Diff/615612765/615634335"

Where is the agreement in this thread for the change? I prefer "found" but can live with "stated". QuackGuru (talk) 05:17, 5 July 2014 (UTC) originally posted at Wikipedia talk:Manual of Style/Words to watch; moved 12:55, 5 July 2014 (UTC)

@QuackGuru: - At the time of posting, the discussion had proceeded only as far as 23:51, 4 July 2014 above, i.e. my and Alexbrn's exchange. I thought his comments made a lot of sense, so using those, I undertook a second round of a WP:BRD cycle. I'd prefer "stated" to "found", as "found" goes beyond "stated" imo, but if you & others are OK with "described", I think that's better prose. IMO, "An editorial in Nature characterized TCM as largely pseudoscience" is a lot better than "An editorial in Nature stated that TCM is largely pseudoscience...". I'm not looking for anything different in meaning from "stated". And if "stated" is what people want, "stated" it will remain. --Middle 8 (leave me alone • talk to me • COI?) 13:16, 5 July 2014 (UTC)

On inspection, this isn't primarily a dispute over word choice, but about neutrality: to what extent does Wikipedia assert that acupuncture is pseudoscience (flavouring it so with words such as "found" or "noted") or suggest more that this is an opinion ("characterized", "described", etc.). I suggest that this perma-discussion is better continued on the Acupuncture article's Talk page. Alexbrn ^{talk|contribs|COI} 05:41, 5 July 2014 (UTC) originally posted at Wikipedia talk:Manual of Style/Words to watch; moved 12:55, 5 July 2014 (UTC)

@Alexbrn: - as far as assertion as fact goes, you may remember our discussion about demarcation aways back, cf. my reply to RexxS below. As far as word choice goes, my intent was to stay neutral, hence posting here. --Middle 8 (leave me alone • talk to me • COI?) 13:21, 5 July 2014 (UTC)

My reasons for reverting were clear in the edit summary. The difference between "stated" and "characterised" is not insignificant, the source did not characterise, it stated. The result, before my revert, allowed a little more wiggle room in the meaning of the sentence. Apart from improving the article, I just made a small repair to the dam, like putting a hand over a hole, just to help stem the tide of fringe pushing. -Roxy the dog (resonate) 07:51, 5 July 2014 (UTC) originally posted at Wikipedia talk:Manual of Style/Words to watch; moved 12:55, 5 July 2014 (UTC)

@Roxy the dog: Your edit summary, "Reverted non neutral change characterised as neutral", [87] and subsequent discussion (see [88]) revealed that your concern was NPOV. Next time if you give me a little more detail, I'll have something to work with and can do a proper WP:BRD cycle. Anyway, Alexbrn helped me surmise that the concern might be about wiggle room, hence "described", [89] if people are OK with that. And FFS, man, AGF a bit; this wasn't a POV push, just a prose improvement, as promised. --Middle 8 (leave me alone • talk to me • COI?) 13:21, 5 July 2014 (UTC)

For me, the distinction is a reflection of WP:ASF: assert simple facts. When an authoritative source (in this case an editorial in Nature) makes a statement that is not contradicted by another equally good source, we should be treating it as fact and reporting it in Wikipedia's voice, like this: "X is true". When there is significant disagreement between sources, then we neutrally report the opinions of the sources by attributing those opinions, like this: "A states X is true; but B states X is false". Sometimes we end up with "A states that X is true", which informs the reader within the prose of the source of that conclusion (even though the citation is easy enough to click on), but starts to mimic the construction we use for disputed conclusions. When the dispute is only among the editors, not among the reliable sources, I'd prefer to stick with ASF. A further change of "A stated X is Y", ("found" or "concluded" are useful alternatives for stylistic variation) to "A described X as Y" (or "characterised") emphasises the author rather than the conclusion and casts the statement even more as an opinion. --RexxS (talk) 10:48, 5 July 2014 (UTC) originally posted at Wikipedia talk:Manual of Style/Words to watch; moved 12:55, 5 July 2014 (UTC)

@RexxS: -- WP's guideline/policy (it's wobbled between both, like RS <==> V) on demarcation is WP:FRINGE/PS. It says "Proposals which are generally considered pseudoscience by the scientific community, such as astrology, may properly contain that information and may be categorized as pseudoscience" and "Hypotheses which have a substantial following but which critics describe as pseudoscience, may contain information to that effect; however it should not be described as unambiguously pseudoscientific while a reasonable amount of academic debate still exists on this point." Obviously we follow RS, and sufficiently weighted RS don't call a topic pseudoscience, then we don't. (A higher threshold -- presence of a source meeting WP:RS/AC calling a topic pseudoscience -- has also been accepted in some topics. But for this discussion I won't apply it.) So we generally follow more than one RS. Acupuncture is considered to be a borderline case, rather than outright pseudoscience or science, by Michael Shermer, in a book coedited by Massimo Pigliucci (Google Books source). There are others, more calling it pseudo than borderline, but enough of the latter to miss the "generally considered" threshold -- hence attribution of the Nature source. But again, I'm not looking for any connotation different from "stated". I hope "described" is that simply because "An editorial in Nature characterized TCM as largely pseudoscience" is a lot better than "An editorial in Nature stated that TCM is largely pseudoscience...". --Middle 8 (leave me alone • talk to me • COI?) 13:18, 5 July 2014 (UTC)

(edit conflict) The first question to ask is "why are attributing to Nature, rather than stating the fact?" If we are basing our content solely on the Nature source, without any other equally reliable source that contradicts it, then we should be reflecting the conclusions of that source as fact, not as merely an opinion. As far as I can see, the source concludes that TCM is largely pseudoscience, with no valid mechanism of action for the majority of its treatments. I'm puzzled why we didn't simply write that? --RexxS (talk) 13:24, 5 July 2014 (UTC)

I think it's clear that neither Nature nor Wikipedia is describing acupuncture as "unambiguously pseudoscientific", despite little dispute that meridians, qi and acupuncture points are about as scientific as astrology. Michael Shermer has created a rather unique triple system of classifications - that of scientific, pseudoscientific, and borderline that isn't reflected in the mainstream. And that's the key point - it's the mainstream scientific viewpoint that determines whether a topic is science or pseudoscience. An editorial from Nature is going to carry a lot more weight when evaluating the mainstream view than a book from a couple of authors - particularly when those authors believe acupuncture is in the same category as string theory.

This description in WP:FRINGE/PS is pertinent:

• "To determine whether something is pseudoscientific or merely an alternative theoretical formulation, consider this: Alternative theoretical formulations generally tweak things on the frontiers of science, or deal with strong, puzzling evidence—which is difficult to explain away—in an effort to create a model that better explains reality. Pseudoscience generally proposes changes in the basic laws of nature to allow some phenomenon which the supporters want to believe occurs, but lack the strong scientific evidence or rigour that would justify such major changes."

Which one is closest to what we're looking at here?

If I look at a Google search on acupuncture pseudoscience, I get 80K ghits and browsing through them, they all seem to equate acupuncture with pseudoscience. On the other hand, searching on acupuncture "not pseudoscience" gives me 1.2K ghits. That's not proof, of course, but it does indicate the difficulty of trying to argue that the mainstream scientific view is 'acupuncture is not pseudoscience'. If you want to attribute Nature, then you will need to provide and attribute the contradictory source without breaching WP:DUE. Frankly I don't think anything along the lines of:

• An editorial in Nature stated that TCM is largely pseudoscience,[ref] but Michael Shermer stated that acupuncture is borderline science.[ref]

is likely to find consensus as being a balanced representation of the scientific mainstream. YMMV, of course. --RexxS (talk) 14:20, 5 July 2014 (UTC)

@RexxS: Again, in rough order of your points:

1. See WP:ASSERT and [Demarcation problem]]. Demarcation isn't an objective fact. Also, FWIW, see my argument at Talk:Acupuncture/Archive_13#More_re_TCM_.26_pseudoscience_wording. But the main point is that all demarcation as pseudoscience is opinion depending on the demarcation criteria used. It's just that some opinions are near-universally held and others aren't.
2. If we say "TCM is pseudoscience" that's precisely the same as describing acupuncture as "unambiguously pseudoscientific". (WP:ASSERT again.) Which we don't do if acu doesn't meet the threshold for "generally considered"; the choice is between that and "questionable science", not "alternative theoretical formulations".
3. That "reasonable debate" over acu's demarcation is evident from sources like Shermer, who represent what WP would call a "significant minority view". His reasons for calling acu borderline include more than just qi and meridians; they include testability or falsifiability, which is Karl Popper's main criterion. Shermer and Pigliucci are well-respected thinkers and likely bigger experts on the demarcation problem than the staff at Nature, unless the latter includes specialists in the philosophy and history of science. "Borderline" is a classification that can be arrived at via different means -- i.e. by answering a variety of differrnt questions. Being in the same category as string theory is no more unusual than a person with mild ADHD being in the same category (i.e. mental illness) as someone with intractable psychosis.
4. We don't have to mention Shermer in the same breath as Nature' we just have to decide whether or not acu fits our "generally considered" category, and then apply WP:ASSERT to how we cite Nature. Just because it describes acu as pseudoscience doesn't mean we have to follow; rather we consider the range of sources.
5. Sure nobody considers acu "not pseudoscience", but some do put it in a 3rd, "grey area" category the way Shermer does. Try a search -- including Google Scholar, where you'll find stronger RS's than in a normal search --for e.g. acupuncture demarcation borderline OR "grey area" . The more scholarly sources are more likely to say that demarcation for acu is fuzzy, and . --Middle 8 (leave me alone • talk to me • COI?) 15:42, 5 July 2014 (UTC)

It's true that 'scientific fact' is someone's opinion in the final analysis, but when it is near-universally held then common sense would demand that we state that mainstream opinion as fact - which is not to say that we can't talk about pseudoscientific or fringe theories, but that has to be in context of the mainstream view. WP:PROFRINGE refers.

But we don't say "TCM is pseudoscience"; we say TCM is largely pseudoscience, with no valid mechanism of action for the majority of its treatments. and that is fully supported by a very high quality citation. Why wouldn't the reader draw the conclusion that acupuncture is largely pseudoscience, rather than "unambiguously pseudoscientific"? Surely we are making it clear in the article (or lead) as a whole that there is some evidence of efficacy - even if it might be no greater than placebo - but that the claimed mechanism, rooted in TCM, is pure mumbo-jumbo?

I'm not sure I'd see Shermer as a significant minority view. Insisting on three categories of demarcation looks more like a minority view of one to me. If it's important to debate demarcation, I can quickly adduce a contrary view from experts on the issue: Massimo Pigliucci and Maarten Boudry (the editors of "Philosophy of Pseudoscience: Reconsidering the Demarcation Problem") in the NYT.

I've got to disagree that string theory and acupuncture belong in the same boat. The difference can be seen by reference once more to WP:FRINGE/PS:

• Alternative theoretical formulations from within the scientific community are not pseudoscience, but part of the scientific process. They should not be classified as pseudoscience but should still be put into context with respect to the mainstream perspective. Such theoretical formulations may fail to explain some aspect of reality, but, should they succeed in doing so, will usually be rapidly accepted.

That's a description of string theory, but acupuncture not only fails to explain reality in any testable way, it relies on a theory that has absolutely no relationship whatsoever to any scientific mechanism. --RexxS (talk) 16:55, 5 July 2014 (UTC)

I am so fed up with pov pushing COI SPA editors chip chip chipping away at good sensible mainstream articles, and when they have the absolute gall to say something like "And FFS, man, AGF a bit; this wasn't a POV push ..." when their whole raison d'etre here is a pov push, I suppose one can either laugh or cry. I sprayed tea all over my monitor. FFS. -Roxy the dog (resonate) 14:35, 5 July 2014 (UTC)

@Roxy the dog: Your comments are out of line. Read the section just above, Talk:Acupuncture#Commenting_about_other_editors. This isn't the place. Time to take this back to your user talk. --Middle 8 (leave me alone • talk to me • COI?)

I like RexxS's proposal, except I'd just change "but" to "and", since the two statements (from Nature & from Shermer) are not incompatible. Alexbrn ^{talk|contribs|COI} 14:46, 5 July 2014 (UTC)

No strong feelings about but vs. and, but since Shermer has three categories and one of the is pseudo which he doesn't use, doesn't that make him and Nature incompatible? --Middle 8 (leave me alone • talk to me • COI?) 14:56, 5 July 2014 (UTC)

No, since Nature say it's largely pseudoscience. Alexbrn ^{talk|contribs|COI} 15:12, 5 July 2014 (UTC)

(Add) Since our readers won't know what Shermer means by "borderland science" (N.B. it is not "borderline", which implies something rather different) it is probably also worth explaining that by this he means an area "between science and nonscience" and that on a scale of 1 to 9 for scientific validity (1 is worst), he assigns acupuncture a score of 3 (i.e. less valid than chiropractic but more than Omega Point Theory). Alexbrn ^{talk|contribs|COI} 15:21, 5 July 2014 (UTC)

@Alexbrn: Yes, I know he said "borderlands(s)" (I just used "borderline" here because it sounds less weird, but the important thing is that it's three categories, whatever they're called). On p. 206 he uses "borderland(s)" as a third category along with science and pseudoscience, not nonscience. Where does he score things? Is there a different source? I'm not seeing it in the chapter in Pigliucci, but maybe Google is blanking pages or something. Re "largely", it's likely not incompatible with Shermer if he gives acu a 3. Curious to see the source though. --Middle 8 (leave me alone • talk to me • COI?) 09:50, 6 July 2014 (UTC)

It's in his OUP book. See:

• Michael Shermer (17 May 2001). The Borderlands of Science : Where Sense Meets Nonsense: Where Sense Meets Nonsense. Oxford University Press. pp. 24–. ISBN 978-0-19-803272-4.

The distinction between borderland and borderline is quite important. "Borderline" implies Shermer sees acupuncture as just on the edge of science, but in fact he places it firmly in a zone outside science (though not in nonscience either) - a sort of non man's land. Alexbrn ^{talk|contribs|COI} 10:30, 6 July 2014 (UTC)

Arbitrary break

About the assertion that there is no source to contradict the Nature editorial - of course there is. There are plenty of sources saying TCM is this or that kind of system. Paul Unschuld, who is an expert but not a proponent or supporter of TCM, defines it as a system of healing that relies upon an elaborate system of correspondences (Medicine in China). Some issues with the Nature editorial include 1. Its an editorial, not the findings of a directed inquiry into the question of "what is TCM". 2. Within that OPINION piece, the text in question is offered as an off-hand dismissal, as in "why hasn't this researched panned out? Maybe because it is largely pseudoscience". With such a weak source, don't you see how its problematic to assert that statement as a fact? IF you must make a ridiculously over-simplified statement like "TCM is …", wouldn't you want to base it on a source like Unschuld, who is an acknowledged expert on the subject but has no reason to see it painted in either a positive or negative light? Herbxue (talk) 17:16, 5 July 2014 (UTC)

I thought it was a given that when we describe claims of a biomedical nature, we use only sources that meet the standards of WP:MEDRS. Are you contending that PMID 17625521 doesn't meet MEDRS standards? Or are you claiming that your unnamed "plenty of sources" all meet those requirements? From what I can see, Paul Unschuld's expertise is in the field of the history of Chinese medicine - do you mean his writing in ISBN 978-0520062160 (1992) should be regarded then as "books written by experts in the relevant field and from a respected publisher"? It's a bit old, but worthy of consideration, of course.

If you insist that "TCM is largely pseudoscience, with no valid mechanism of action for the majority of its treatments" is a ridiculous over-simplification, then let's just settle it with an RfC. I suggest the question asked should be:

• Is "TCM is largely pseudoscience, with no valid mechanism of action for the majority of its treatments" a ridiculous over-simplification? Yes or No?

Feel free to suggest an alternate equivalent formulation. --RexxS (talk) 18:03, 5 July 2014 (UTC)

I think you are dismissing my concerns about the way the source is being used without actually addressing the specific concerns I brought up. I already settled for the compromise of in-text attribution, and would be happy to leave it that way. But the opinion in the editorial should not be presented as established biomedical fact (for the reasons I mentioned). Herbxue (talk) 19:40, 5 July 2014 (UTC)

Btw I think it does fail MEDRS - something more formal than one (or several) writer's opinion should be required to establish fact in an article like this. For all the descriptive info (definitions of Qi, Yin, Yang, etc.) a simple textbook should suffice as all sources generally agree on what those definitions are in the context of TCM. For a sweeping general statement about what TCM IS (medical system? folk healing tradition? pro to-science? pseudoscience? ) I think we need something more robust.Herbxue (talk) 19:46, 5 July 2014 (UTC)

I agree Unschuld may be an important voice here. But in truth his position, at least recently, is properly skeptical – so much so that he was recently attacked in the fringe journal Journal of Chinese Medicine in a piece entitled "Traditional Chinese Medicine - Science or Pseudoscience? A Response to Paul Unschuld". It said,

It appears that Unschuld characterises Chinese medical theories as 'magical' - i.e. pseudoscientific - thinking. This article examines the tacit beliefs which appear to underlie the work of Professor Unschuld (and that seem to be shared by other prominent authors such as Joseph Needham and Ted Kaptchuk) that deny Chinese medicine equal status with modern biomedicine ...

. Alexbrn ^{talk|contribs|COI} 19:55, 5 July 2014 (UTC)

(edit conflict) @Herbxue: You mean you want me to take your amateur analysis of a MEDRS-compliant source as a reason to doubt it? It's not our job - and we're not qualified - to do that. Here's what the source says about pseudoscience:

• So if traditional Chinese medicine is so great, why hasn't the qualitative study of its outcomes opened the door to a flood of cures? The most obvious answer is that it actually has little to offer: it is largely just pseudoscience, with no rational mechanism of action for most of its therapies. ... But it seems problematic to apply a brand new technique, largely untested in the clinic, to test the veracity of traditional Chinese medicine, when the field is so fraught with pseudoscience.

You are now claiming that

• "TCM is largely pseudoscience, with no valid mechanism of action for the majority of its treatments"

is a misuse of the source - as well as "ridiculously over-simplified". I think you're completely wrong, but I'd be quite happy to hear more opinions on the issue. I'll drop a note at WT:WPMED to see if we can get more views. --RexxS (talk) 20:03, 5 July 2014 (UTC)

Additionally, as you think "Hard to Swallow" fails MEDRS, I've asked at WP:RSN for further opinions on whether that is the case. --RexxS (talk) 20:29, 5 July 2014 (UTC)

@Alex - thanks for pointing that out. It shows that the acupuncture profession are not exactly "fans" of Unschuld, but those of us that teach the subject and are interested in understanding its origins regard him as a high authority, even if he believes TCM is pseudoscience. Now, if he states that directly in an article or book, I would not contest it (unless it were an EDITORIAL, in which case I would recommend in-text attribution. @Rex - yes more opinions are of course welcome, but you directly disrespecting me rather than address the specific concern is not. Did you not get the memo? There is a difference between what the rules say you CAN do and what we, together, should do to improve the article. Wish you would try to come to consensus with me instead of insulting me.Herbxue (talk) 20:37, 5 July 2014 (UTC)

• Support the wording "TCM is largely pseudoscience, with no valid mechanism of action for the majority of its treatments" is supported by many sources and is the main stream POV. Here is another Nature article, this one from 2011, which states "The local Asian one is based on traditional Chinese medicine (TCM) — herbal mixtures developed though observation and experience accumulated over thousands of years, but with unknown mechanisms of action" thus supporting the "no valid mechanism of action" and "TCM still have their firm defenders, such as Boli Zhang, president of the China Academy of Chinese Medical Sciences (CACMS) in Beijing. “We believe in the jing luo [the meridian or energy pathways], but we have not found it yet,” he says. Nevertheless, even in the TCM community, there are fewer Chinese scholars who believe the TCM tenets literally. Indeed, there has been criticism from academics and the media in China, arguing that much of TCM and most of its theories are pseudoscience and that China should bid “farewell to traditional Chinese medicine”" supporting the pseudoscience bit [90] If there are no good sources that state there is a valid mechanism of action I support Rexx's summary. Doc James (talk · contribs · email) (if I write on your page reply on mine) 20:56, 5 July 2014 (UTC)

And than we have books like this [91] from 2013 that more or less say the same. The Nature articles however are sufficient. Doc James (talk · contribs · email) (if I write on your page reply on mine) 21:03, 5 July 2014 (UTC)

If the Nature article were enough (the other one does not say the same thing) why bring up the others? Unknown mechanism does not equal "no mechanism". The statement is an opinion. Even if it is the most popular opinion, it does not make it fact. If you want a good article, why insist on dismissive wording from an editorial? We can do a little better than that. The issue has more nuance, and in-text attribution still includes the text but makes sure the reader knows that it is not the unanimous opinion of the medical profession (despite it seeming like that that is an obvious and acceptable assumption). It is true that even within the profession, ideas like the 5 phase theory are sometimes referred to as "naive", but there is no unanimous position. I don't believe "science" or "medicine" have unanimous opinions either.

Also, failing to find a source saying there is a valid mechanism of action is a justification for calling it pseudoscience is original research. Herbxue (talk) 21:15, 5 July 2014 (UTC)

It is not just one high quality source that supports this but many. And as we have high quality sources that call it pseudoscience it it is not OR. Here is a book from ACP press that says with respect to acupuncture / TCM that their is an "absence of a logical underlying medhcanisms of action" [92].

This textbook says "a plausible mechanism of action is not available" [93] I am happy with Rexx's wording and am happy to have a RfC regarding this question to bring in more input if people wish. Doc James (talk · contribs · email) (if I write on your page reply on mine) 21:37, 5 July 2014 (UTC)

Those sources support the mechanism text but not the pseudoscience wording. Pseudoscience is something that presents itself as a science. I think bringing in more voices is probably a good idea. Rexxs posted the question a couple places already.Herbxue (talk) 21:49, 5 July 2014 (UTC)

So is your position that acupuncture does not present itself as science? If so what does it present itself as? Doc James (talk · contribs · email) (if I write on your page reply on mine) 22:00, 5 July 2014 (UTC)

Traditional medicine, based mostly on theories and practices described in the Yellow Emperor's Classic, written around 2200 years ago. While the schools now teach a robust biomedical curriculum, and there is more and more push towards "evidence-informed practice", treatments are still mostly based on traditional diagnostic skills and "pattern differentiation" An exception is orthopedic cases, in which knowledge of trigger and motor points is commonly taught. Even while familiar with which herbs are antimicrobial, practitioners will still use traditional combinations based on the pattern of symptoms and signs, as traditionally described. Herbxue (talk) 00:03, 6 July 2014 (UTC)

I saw the note at WT:MED about this.

I see two halves of this sentence, and I have very different reactions to them:

"TCM is largely pseudoscience"—Really? For something to be pseudoscience, it must try to present itself as being modern science. I was under the impression that TCM, which is more than two thousand years old, did not pretend to be "scientific". I thought it claimed to be traditional, i.e., to have existed for more than twice as many centuries as the modern scientific method. If you don't claim that your thing is scientific, then you cannot be pseudoscientific (by definition); you can only be wrong. To give a Western example, Hippocrates, the father of Western medicine, was really into this humorism thing. His theory was entirely wrong, but it's not pseudoscience, because he never even pretended that it met modern scientific standards.

"with no valid mechanism of action for the majority of its treatments"—Sure, that's probably true, at least to a first approximation. WhatamIdoing (talk) 04:01, 6 July 2014 (UTC)

Fully agree with WhatamIdoing. Part perhaps of a broader viewpoint problem that Wikipedia has, plausibly related to the undue effects of POV battles such as this one. Prescientific thought is historically and culturally significant encyclopedic content. But if readers look for "Prescientific" or "Prescientific system", they get anachronistically redirected straight to Pseudoscience. Not good, imo. 109.156.204.159 (talk) 09:59, 6 July 2014 (UTC)

It looks like those were victims of a temporary redirecting of Protoscience in 2011 (to which they were pointed as a result of a 2007 AFD). The bot resolves double redirects, but it doesn't undo them when the blanking/redirecting is undone. WhatamIdoing (talk) 23:34, 6 July 2014 (UTC)

Interesting. Thanks for sorting that :) 109.156.204.159 (talk) 13:20, 7 July 2014 (UTC)

(I cross-posted this at RSN --Middle 8 (leave me alone • talk to me • COI?))

I sense an eagerness among editors to label TCM as pseudoscience that is not matched by the Nature source, whose topic is drug discovery and TCM. In context, the quote (below) isn't that strong of an assertion.

To get a sense of the editorial's emphasis, just look at each of its six paragraphs: 1 & 2 = Pharmaceutical interest in TCM. 3 = limited success so far from standard "reductionist" (their word) approaches. 4 = why no success? & quote which we cite (below). 5 = economic & regulatory environment. 6 = non-standard ways of studying TCM formulas; but they're skeptical of this; see 2nd quote below.

Now here's the full quote for our citation (para #4):

"So if traditional Chinese medicine is so great, why hasn't the qualitative study of its outcomes opened the door to a flood of cures? The most obvious answer is that it actually has little to offer: it is largely just pseudoscience, with no rational mechanism of action for most of its therapies. Advocates respond by claiming that researchers are missing aspects of the art, notably the interactions between different ingredients in traditional therapies."[94]

That's more of a point-counterpoint thing, although it's clear which side they're taking. Still -- as Jytdog says -- not the strongest thing upon which to hang WP's voice.

Later in the editorial they say (para #6):

"Constructive approaches to divining the potential usefulness of traditional therapies are to be welcomed. But it seems problematic to apply a brand new technique, largely untested in the clinic, to test the veracity of traditional Chinese medicine, when the field is so fraught with pseudoscience. In the meantime, claims made on behalf of an uncharted body of knowledge should be treated with the customary scepticism that is the bedrock of both science and medicine."[95]

So again, it's obvious where they stand -- but again, given the content, it's not the central point of the paper.

I'm sure there are other sources that call TCM pseudoscience, but they haven't been proposed as a source for speaking in WP's voice, perhaps because they're not as prestigious as Nature. I sense an overreaching among editors, a grasping at whatever is most likely to justify calling TCM pseudoscience -- but these things take time and the literature has not reached the point homeopathy did some time ago. Likewise, the skeptical blogosphere's eagerness to label TCM pseudoscience is not matched by the same enthusiasm among scientists who study TCM treatments.

The best sources would be those most qualified to comment on demarcation: scholars, including historians and philosophers of science, as well as scientific academies who periodically make such statements to educate the public. --Middle 8 (leave me alone • talk to me • COI?) 09:02, 6 July 2014 (UTC)

Would going with "TCM is largely not scientific" address the issue? Doc James (talk · contribs · email) (if I write on your page reply on mine) 03:52, 7 July 2014 (UTC)

The source does not specifically say it is not scientific. See "The most obvious answer is that it actually has little to offer: it is largely just pseudoscience, with no rational mechanism of action for most of its therapies."[96] QuackGuru (talk) 04:06, 7 July 2014 (UTC)

Agree the source does say it is pseudoscience. Some within TCM claim TCM to be science thus making pseudoscience possible. Some however do not claim it is science thus making it "not scientific"

We have a WHO document that states "But they also held that Chinese medicine was not scientific, with their central argument being “Though Chinese medicine can cure diseases, but nobody can explain the logic principles of curing diseases to the people”. [97] So TCM could be both. Doc James (talk · contribs · email) (if I write on your page reply on mine) 04:29, 7 July 2014 (UTC)

@Jmh649:/Doc James: Saying "TCM is largely not scientific" is one of the better ideas in this topic idea that I've heard in awhile. --Middle 8 (leave me alone • talk to me • COI?) 06:32, 7 July 2014 (UTC)

Per WP:PSCI, we should state what is pseudoscience. I could not verify the Nature source says "TCM is largely not scientific". QuackGuru (talk) 06:42, 7 July 2014 (UTC)

that's because we're trying to summarize more than one source. --Middle 8 (leave me alone • talk to me • COI?) 08:18, 7 July 2014 (UTC)

Irrespective of what that particular Nature editorial says, as an encyclopedia we need to maintain an editorial sense of historical and cultural perspective. Stating that a tradition as ancient as acupuncture (or other traditional Chinese medical arts) is a "pseudoscience" can only really make sense within the context of claims made within our own scientific era. The history of acupuncture stretches back to long before modern science, and is rooted in a largely different cultural tradition to that of Western natural science. 109.156.204.159 (talk) 13:17, 7 July 2014 (UTC)

And so does the tradition of telling stories about leprechauns and fairies. Who are we to call them unreal? Tgeorgescu (talk) 13:23, 7 July 2014 (UTC)

My comment was clearly an editorial one (have a look at the Wikipedia pages on leprechaun and fairy). 109.156.204.159 (talk) 13:48, 7 July 2014 (UTC)

WP:MEDRS and WP:MEDASSESS privilege reputable peer-reviewed research (corresponding to modern science) over all other cultural and historical views. These are the policies to be followed in every Wikipedia medical article. Tgeorgescu (talk) 18:53, 7 July 2014 (UTC)

The guideline/s you mention (which I am a strong supporter of and have contributed to in minor ways, though they were largely compiled by experienced Med editors, including WhatamIdoing) primarily regard the sourcing of biomedical claims, as distinct from other legitimately encyclopedic historical and cultural topics, such as fox spirits, etc. Saying that many biomedical claims emanating from the TCM community have been characterized as "pseudoscience" (broadly per the Nature editorial—whether that is deemed an appropriate MEDRS here or not) is quite a different kettle of fish from characterizing the entire tradition, with its long cultural history, as "pseudoscience". That would clearly be anachronistic on historical grounds, and I don't think that particular way of framing the overall subject matter could genuinely be justified by any current Wikipedia policy or guideline.

tl:dr - As editors we need to watch our words and be careful how we frame what we report. 109.156.204.159 (talk) 20:24, 7 July 2014 (UTC)

You wrote "that's because we're trying to summarize more than one source."

Summarizing different sources to come to a new conclusion is SYN. QuackGuru (talk) 17:22, 7 July 2014 (UTC)

True, but summarizing different sources to come to the same conclusion as the sources is called "good editing".

I strongly prefer the WHO document over an unsigned, six-paragraph opinion piece (which doesn't appear to be findable on PubMed, by the way). It's consistent with many other sources, too. This scathing magazine opinion calls TCM "superstition" and "not science". This piece by an acupuncturist says, "Oriental medicine as it is practiced is not scientific medicine. It is something very different, certainly a medical art, but certainly not dependent upon or based upon modern science.Oriental medicine as it is practiced is not scientific medicine. It is something very different, certainly a medical art, but certainly not dependent upon or based upon modern science". Here's an article from People Daily: "However, there is a widespread disenchantment with TCM, especially in medical circles. These doctors believe that TCM is not scientific." This view is presented, in the exact words "not scientific", in many sources. There is no policy violation in accurately representing this widespread view. WhatamIdoing (talk) 23:31, 7 July 2014 (UTC)

But we should still state was is pseudoscience according to policy. Do you also want to include "not scientific" in the article? I think we should use the Nature source which is a highly respected journal. QuackGuru (talk) 23:41, 7 July 2014 (UTC)

No. There is no policy that says, "If you can find a single, solitary source calling something pseudoscience, even if most sources don't choose that term, and even if some sources explicitly reject that term, then you must say that it is 'pseudoscience'." There's no such policy. We are not required to include that word merely because you found one source that uses it.

We're supposed to use DUE weight, which means writing what most sources say, not what a single unsigned editorial opinion—an opinion almost certainly authored by a non-expert—says. Most reliable sources don't choose to describe TCM as pseudoscience. We should use the words that most reliable sources use. WhatamIdoing (talk) 22:25, 8 July 2014 (UTC)

There a WP:DRN for the TCM page and the result was to include pseudoscience in both the lede and body. QuackGuru (talk) 00:45, 9 July 2014 (UTC)

Wording in regards to the Moffet reference

Hi! The original phrasing "To that end, a systematic review found no difference between "true" acupuncture and "sham" acupuncture" was changed to the following: "There is no difference in effectiveness between "true" acupuncture and "sham" acupuncture"[98]. If you ask me, that's not in line with the statistical expression and gives a faulty picture about the source. The source does not say there is no difference, but instead "...found no statistically significant difference in outcomes".

"No difference" and "no statistically significant difference" have a whole different meaning in statistical sense. I'd suggest that the sentence to be phrased as: "A systematic review found no statistically significant difference between "true" acupuncture and "sham" acupuncture...."

That would be also consistent with the phrasing at Acupuncture#Peripheral osteoarthritis. Cheers! Jayaguru-Shishya (talk) 20:08, 5 July 2014 (UTC)

The current text is a good summary, there is no "whole different meaning". Alexbrn ^{talk|contribs|COI} 20:13, 5 July 2014 (UTC)

Actually there is, what JS is proposing is more responsible (not like the evening news - "now breaking chocolate is good for you" then next week "chocolate is bad for you". Moreover there are other reviews that do show statistically significant difference. Are we going to change the text every time a new systematic review comes out with a different conclusion? It would be more responsible to do the more detailed thing we have been doing: "A 2012 review show…", "A 2013 review showed the opposite by comparing…" I think more detail is better when the results are so mixed between reviews. Herbxue (talk) 20:46, 5 July 2014 (UTC)

No. This isn't a view that changes by the week, but when a new systematic review is published, we update our content to summarise any new conclusions. Why would we do anything else? You keep claiming there are "other reviews" that contradict the sources we use, but you haven't provided anything for us to look at here. Of course when reliable sources reach opposing conclusions, we neutrally report both views and attribute them. Here's what our policy WP:Neutral point of view says:

• Avoid stating seriously contested assertions as facts. If different reliable sources make conflicting assertions about a matter, treat these assertions as opinions rather than facts, and do not present them as direct statements.
• Avoid stating facts as opinions. Uncontested and uncontroversial factual assertions made by reliable sources should normally be directly stated in Wikipedia's voice. Unless a topic specifically deals with a disagreement over otherwise uncontested information, there is no need for specific attribution for the assertion, although it is helpful to add a reference link to the source in support of verifiability. Further, the passage should not be worded in any way that makes it appear to be contested.

I hope that helps.

@Jayaguru-Shishya: Why should the undisputed conclusion of a reliable secondary source have to be attributed? WP:ASF requires us to attribute opinions, not facts, and attributing facts as if they were opinions dilutes the undisputed conclusion of a reliable secondary source to make it seem less authoritative. No thanks.

Are you seriously claiming that there is a "whole different meaning" if we write 'no difference' rather than 'no statistically significant difference'? We summarise our sources, not go into nit-pickingly ludicrous detail like that. Just how would our readers understand 'no difference'? If it was significant (statistically or otherwise), it would be 'a difference', not 'no difference'. In addition your formulation fails to explain that any difference is in the context of outcomes ('effectiveness' to translate the medical term in my summary). I'd be happy to make the phrasing at Acupuncture#Peripheral osteoarthritis fit with the consensus that emerges here, but there's no rush. --RexxS (talk) 21:16, 5 July 2014 (UTC)

@RexxS: and everyone: It's not undisputed at all, hence debates like the one Acupuncture#Pain, and the summary in the 2nd para of the lede. To be clear, sham is the control and true is the test group. If there were agreement that they were the same there wouldn't be a debate. But we have reviews finding efficacy, a prominent one being Vickers et. al. 2012: "Our results from high quality trials show that acupuncture is effective for the treatment of chronic pain, with significant differences between true and sham acupuncture indicating that acupuncture is more than a placebo."([99], open-access). Also, see Lee & Done for Cochrane: "Conclusion: P6 acupoint stimulation prevented PONV" ([100]). There's quite an active debate over all this, and it's the single most-discussed topic about acu in the sci literature and the blogosphere, by far. --Middle 8 (leave me alone • talk to me • COI?) 06:54, 6 July 2014 (UTC)

RexxS, you said you are a statistician, right? I assume that you acknowledge, naturally, the difference between "no difference" and "no statistically significant difference". We have varying observations, averages and standard deviations (i.e. they are different). There when you begin testing for statistical significance (no matter what distribution you are resorting to), you will actually see if the difference is statistically significant at the giver risk level. Am I right?

You asked: "Are you seriously claiming that there is a "whole different meaning" if we write 'no difference' rather than 'no statistically significant difference'?" Yepp. if you come to think of it, "no difference" means that all the observations between group A and group B are identical. "Statistically insignificant" doesn't make such bold and absurd assumption. Besides, "no statistical significance" is closer to the source.

I find it pretty absurd to be teaching this to someone who says to have been teaching statistics for 20 years, according to your own words. Let's take an example: "A recent survey on wages stated that the average hourly wage of men was 10 dollars, whereas the average hourly wage of women was only 8,4 dollars. The difference, however, was not statistically significant". Now, according to your logic, you will say: "there was no difference". Ehem, yes there was. But due to other variables, possibly the standard deviation and sample size, even though the difference existed, it was not statistically significant.

You also said: "WP:ASF requires us to attribute opinions, not facts, and attributing facts as if they were opinions dilutes the undisputed conclusion of a reliable secondary source to make it seem less authoritative. No thanks."

??? The source says "no statistically significant difference", that's not an opinion. Nor it's an absoulte fact. It's just an outcome according to one study (systematic review). Jayaguru-Shishya (talk) 22:41, 5 July 2014 (UTC)

Wikipedia is an encyclopedia written for the general reader, not just for statisticians. I don't need a lecture from you on the use and abuse of statistics as I'm comfortable enough with statistical methods to know when it's appropriate to go into detail about statistics and when it's not. If two outcomes are not statistically different, then why would we write anything other than there is no difference in the outcomes? Do you think there is a difference in the effectiveness of "true" and "sham" acupuncture? If so, then what do you base your opinion on? If not, then we agree there's no difference in effectiveness and you're just arguing a pedantic point tendentiously. --RexxS (talk) 23:20, 5 July 2014 (UTC)

You certainly know that the difference in meaning is important, and your statement about your authority on stats doesn't help. "no difference" implies that the two phenomena are to be considered to be the same. What we have here is one review in which no statistically significant difference in efficacy was found. To generalize that into "there is no difference between true and sham" is a reach.Herbxue (talk) 23:35, 5 July 2014 (UTC)

@RexxS:Yes, most certainly there is a difference between sham acupuncture and true acupuncture. That's a fact that I base on the full version of the study by Moffet. I can email it to you if you like. Well, you don't need the full one even; the abstract already says that 22 out of 38 trials found no statistically significant difference (while the rest 16 trials naturally did).

When we say "no difference", it means that the observations in our data are identical. When we are testing for statistical significance though, we have different estimates and we are looking for the probability that the differences might be due to randomness at the given risk level (0.05; 0.01; 0.001; whatever you like). I hope you are not teaching you students that "difference" and "statistical difference" are the same ;) Jayaguru-Shishya (talk) 12:58, 6 July 2014 (UTC)

When we use the results of a statistical test to say there is no difference in a conclusion, we mean that the null hypothesis "there is no difference in this conclusion between the given data sets" is probable within a given certainty (often 95% or 99%). That doesn't mean that the data sets are identical - how ludicrous is that assertion. If I toss two coins 100 times each and one comes down heads 50 times and the other 49 times, I write that they are equally likely to be unbiased coins; I don't need to qualify that for the general reader. Although I haven't taught statistics to students for more than 20 years, I hope the students I taught learned how to communicate in plain English and don't try to baffle their audience with spurious statistical qualifiers when they are wholly unnecessary. --RexxS (talk) 18:43, 7 July 2014 (UTC)

Agree "no statistically significant difference" when simplified equals "no difference" Doc James (talk · contribs · email) (if I write on your page reply on mine) 23:36, 5 July 2014 (UTC)

Actually it's the review that examined the statistics in the primary studies that concludes "true" and "sham" acupuncture have the same outcomes. So tell me, Herbxue, do you believe there's a difference between the outcomes from "true" and "sham" acupuncture? Feel free to actually quote a source to justify your claims for a change. --RexxS (talk) 23:50, 5 July 2014 (UTC)

I highly, highly recommend that all editors review the article, especially the role of sham acu, and sources' varying conclusions on efficacy, before making further bold edits along these lines. And our sources. Moffet shouldn't have been cited in the first place for "there is no difference"; that was never a conclusion for the whole dataset (let alone the entire literature!). Then a debate erupted over wording in the article that never should have been there. --Middle 8 (leave me alone • talk to me • COI?) 10:22, 6 July 2014 (UTC)

Wording not supported by source (let alone rest of literature)

The wording "There is no difference between "true" acupuncture and "sham" acupuncture" ([101][102]) isn't supported by the source, nor by sci consensus on the literature. Moffet (abstract; full paper paywalled) is titled: "Sham Acupuncture May Be as Efficacious as True Acupuncture: A Systematic Review of Clinical Trials". Findings: about 60% of papers reviewed found no significant difference between sham and true acu. Conclusion: "The findings cast doubt on the validity of traditional acupuncture theories about point locations and indications. Scientific rationales for acupuncture trials are needed to define valid controls, and the theoretical basis for traditional acupuncture practice needs to be re-evaluated." We can cite it for "...the theoretical basis.." language but not the "There is no difference....".

Sham acu is a control, so if it's a general truth that sham acu is the same as true acu, then acu is not effective -- full stop. That's the crux of the whole efficacy debate, and per other sources in the article, that's not resolved, especially for nausea and pain. See other stuff in lede and body for that. The wording for Moffet was inaccurate even before the edit that removed attribution wording. -Middle 8 (leave me alone • talk to me • COI?) 03:46, 6 July 2014 (UTC)

I reverted the attribution wording and changed the "there is no difference..." wording to reflect the source (diff). New wording, using actual title of paper: "To that end, a systematic review found that "sham" acupuncture may be as efficacious as "true" acupuncture, therefore, the validity of traditional acupuncture theories including acupuncture point locations has been questioned." --Middle 8 (leave me alone • talk to me • COI?) 04:22, 6 July 2014 (UTC)

User:Middle, you made changes against a clear WP:CON. You unilaterally restored the old wording that was previously rejected. The previous wording was discussed in detail at Talk:Acupuncture#Original research again and the previous wording you changed is supported by the source. Doc James made the specific proposal on the talk page.[103] I made the proposal in mainspace and this was also supported by another editor[104]. QuackGuru (talk) 02:07, 7 July 2014 (UTC)

Issues with the change by User:Middle 8 to "found that "sham" acupuncture may be as efficacious as "true" acupuncture" is that it is exactly the same as the source which says "found that sham acupuncture may be as efficacious as true acupuncture"[105] this this appears to be a copy and paste violation. We must paraphrase sources. This is an absolute requirement. Also unless their is an equally high quality source we do not need to start every sentence with "a systematic review". What was there before also needed more paraphrasing. Doc James (talk · contribs · email) (if I write on your page reply on mine) 04:02, 7 July 2014 (UTC)

Well there is a clear problem with that. If we summarize the Vickers paper and present it in WP's voice, then we summarize this source and present it in WP's voice, we will contradict ourselves. I support being as accurate as possible. Say what source concluded what, then there is no issue. I'm not pushing a pro TCM POV if I am just accurately reporting what was found in a systematic review. Similarly, you will not be pushing an anti-acu POV by being as accurate as possible. If you choose to present one systematic review's conclusion as fact, you have weight problems, OR, and a POV problem. Herbxue (talk) 04:29, 7 July 2014 (UTC)

I am open to the discussion of alternative wording and sources. I am not open to copyright issues. Doc James (talk · contribs · email) (if I write on your page reply on mine) 04:31, 7 July 2014 (UTC)

Glad you changed it, it's better wording now. To be honest I thought that if I didn't use the exact wording, it would be seen, somehow, as a POV-push. My mistake was not putting it in quotes. --Middle 8 (leave me alone • talk to me • COI?) 07:57, 7 July 2014 (UTC)

BTW this business of using WP's voice for high-quality sources is a terrible idea -- one that we need to nip in the bud -- for areas where there is controversy, and/or simply unsettled questions, in the literature. That this is the case in this topic area should be completely obvious from reading the article, esp. para #2 of the lede, and the efficacy section. Equally it should be obvious that the statement "there is no difference between sham acu and true acu" is the exact same thing as saying "the efficacy question is settled and there is no efficacy", and that this statement contradicts the article. --Middle 8 (leave me alone • talk to me • COI?) 10:20, 7 July 2014 (UTC)

You claimed it was "new wording" you added but you restored old wording. Now you claim "To be honest I thought that if I didn't use the exact wording, it would be seen, somehow, as a POV-push". How could old wording be new wording? QuackGuru (talk) 21:20, 7 July 2014 (UTC)

I was requested to review this change on the basis that it constituted a form of edit-warring. I could probably write an explanation of how it was edit-warring and use that against Middle 8. I'm not going to, though, because that would be a misguided application of rules against substance. The problem is the Middle 8 is right that the contents of the source can't be used to support the statement that was in the article. The source is a barely-credible source that everyone would be happy to dismiss if it had come out in favor of acupuncture. Since it came out against, it's very tempting to hold our noses and try to use it as proof that acupuncture is crap. However, the source isn't usable for that: it's hard to judge from the abstract what it even means that 22 out of 38 found no significant difference, and that 13 of those 22 found that it may "equally efficacious". I'm hard put to understand exactly what the 9 that found no significant difference but didn't find them to be equally efficacious found: that they were statistically the same but unequally effective? So, the study reviewed 16 studies that found a significant difference, 13 that found no significant difference, and 9 that found they were statistically the same but unequally effective (which strikes me as an incoherent result). That study can be used to source a lot of things, but a statement in Wikipedia's voice making a flat dismissal of anything but the competence of those writing the study? No. If you want the statement to survive, find a source that isn't so frustratingly vague. For Doc James's benefit: if properly indicated as a quote, quoting the title as the conclusion isn't going to present a copyright issue.—Kww(talk) 04:40, 7 July 2014 (UTC)

It was not in quotes and writing an encyclopedia should not be simply a bunch of quotes joined together. I have not examined the quality of the source to any great degree.

It is a systematic review but the impact factor of the journal is low at 1.46 [106]. Is this journal reputable? Maybe not. Doc James (talk · contribs · email) (if I write on your page reply on mine) 04:47, 7 July 2014 (UTC)

It's a fringe journal. Using it goes against the grain of WP:FRIND, which applies here. Alexbrn ^{talk|contribs|COI} 06:16, 7 July 2014 (UTC)

Have removed. Doc James (talk · contribs · email) (if I write on your page reply on mine) 06:38, 7 July 2014 (UTC)

@Middle 8: the policy of stating uncontested statements from reliable sources in Wikipedia's voice is part of WP:NPOV, one of the three core policies, and is not negotiable. Editing against core policy is disruptive and may lead to an editor losing their privilege to edit here, so please step away from trying to "nip in the bud" attempts to align the content with policy. It should be obvious that "uncontested" means among reliable sources, not between editors, so let's make sure that statements from reliable sources are either:

• made in Wikipedia's voice as fact when they are the only reliable source (or represent the overwhelming majority of sources (WP:YESPOV); or
• attributed as opinion along with the opposing view where both views have currency.

As for the argument that "if it's a general truth that sham acu is the same as true acu, then acu is not effective -- full stop", you seem to have missed the point that both "sham" and "true" acupuncture may be effective simply as placebos - a point made many times in the literature. In fact we read that where the practitioner is very enthusiastic about outcomes, the effectiveness increases significantly (I'll find you the paper if you haven't already seen it). Acupuncture could only be shown to work where it is better than a placebo, and placebos are pretty good with pain generally. I'm not seeing strong evidence from secondary sources that demonstrate that "true" is more effective than "sham", so we should be stating that. If you believe that a secondary source concludes the opposite, then we summarise both and attribute properly (Lee & Ernst 2011 may be helpful there). --RexxS (talk) 15:50, 7 July 2014 (UTC)

@RexxS: The point is that the sources under discussion are contested; please re-read my comment to which you're referring. (Also this and my reply just below, simultaneous with this one.) Re placebo, no, haven't missed a thing there; all my comments take that into account. (It's safe to assume that I understand WP policy and the topic area (both from a TCM and scientific angle) and the scientific method, so no need to recap the basics of these to me, though I appreciate the intention. I'm not much good at statistics though.) --Middle 8 (leave me alone • talk to me • COI?) 07:28, 8 July 2014 (UTC)

But for a source to be contested, the other contesting source has to be supplied. It's not enough to just say "Some other (unnamed) sources contest it" although that's exactly what was happening at the time. I accept that other sources are now being discussed.

If you took into account the possible placebo effect, how could you conclude "if it's a general truth that sham acu is the same as true acu, then acu is not effective" when that is plainly untrue if both were to be effective as placebos? You still haven't corrected yourself on that logical faux-pas - or even admitted to it. --RexxS (talk) 13:13, 8 July 2014 (UTC)

As Middle8 mentioned above "Vickers et. al. 2012: "Our results from high quality trials show that acupuncture is effective for the treatment of chronic pain, with significant differences between true and sham acupuncture indicating that acupuncture is more than a placebo."([102], open-access). Also, see Lee & Done for Cochrane: "Conclusion: P6 acupoint stimulation prevented PONV" ([103]). " I mentioned this source above as well. The problem with stating in WP's voice is that there are contradictory reviews. Even Ernst 2011 mentions the trend towards more positive reviews in recent years. I don't see why you feel the rules are forcing you to reach for more conclusive language than that found in the literature. Why not just be careful and attribute clearly?Herbxue (talk) 16:21, 7 July 2014 (UTC)

I updated the lede using the systematic review of systematic reviews.

The weaker Vickers 2012 source was contradicted. Commenting on this meta-analysis, both Edzard Ernst and David Colquhoun said the results were of negligible clinical significance.[62][63] QuackGuru (talk) 18:20, 7 July 2014 (UTC)

(edit conflict) You mean Acupuncture for chronic pain, Vickers et al (2012)? Which goes on to say "However, these differences are relatively modest, suggesting that factors in addition to the specific effects of needling are important contributors to the therapeutic effects of acupuncture" - not exactly a ringing endorsement for the claim there's a difference in outcomes between "true" and "sham" acupuncture.

Nevertheless, when there's a difference of opinion between equally reliable sources, we should attribute them something like this:

• A 2009 review by Moffet concluded that "sham" acupuncture may be as effective as "true" acupuncture, whereas a 2012 review by Vickers et al found that "true" acupuncture showed a modest, but significant, improvement in effectiveness over "sham" acupuncture.[ref for Moffet][ref for Vickers]

If we were to discard Moffet's review for some good reason, then the content ought to phrased something like this:

• "True" acupuncture shows a modest, but significant, improvement in effectiveness over "sham" acupuncture.[ref for Vickers]

I hope that makes clearer the policy I'm asking us to follow. --RexxS (talk) 18:27, 7 July 2014 (UTC)

@RexxS: there are a lot more sources than just Vickers. There are dozens of systematic reviews. When some aspect of science is in flux, as here with efficacy, we should avoid using WP's voice altogether and just attribute (when speaking of that contested or unsettled aspect of the field). Or if we're really confident about the debate boiling down to a finite number of positions, which it might in this case, we could use WP's voice to state both (or all) sides, but again this risks WP:SYN and we might look to reviews of reviews. (more re placebo: efficacy in this context is understood as meaning efficacy beyond placebo, or specific activity -- IOW the difference between sham and true acu.) Agree we need to distinguish between clinical and statistical significance. Note that that Vickers and Colquhoun have disagreement about clinical significance: Vickers argues that the clinical decision is whether to refer, and so the deciding factor is the sum of specific and non-specific (placebo) effects. Note also that sham acu can mean different things, typically either fake needling at real points, or real needling at non-acupoints. I think we'd all do well to focus less aggressively on the lede and make sure we get this stuff right in the body first. --Middle 8 (leave me alone • talk to me • COI?) 07:28, 8 July 2014 (UTC)

Nevertheless, when there's a difference of opinion between equally reliable sources, we should attribute them something like this:
* A 2009 review by Moffet concluded that "sham" acupuncture may be as effective as "true" acupuncture, whereas a 2012 review by Vickers et al found that "true" acupuncture showed a modest, but significant, improvement in effectiveness over "sham" acupuncture.[ref for Moffet][ref for Vickers]

Yes, exactly. Jayaguru-Shishya (talk) 21:28, 7 July 2014 (UTC)

The lede now says: A systematic review of systematic reviews found that for reducing pain real acupuncture was no better than sham.[5] QuackGuru (talk) 18:33, 7 July 2014 (UTC)

QG's text is almost fine because it is clear where it came from, but it should say "one review from Ernsts review of reviews found no difference between sham and verum for back pain" - your text over-generalizes the finding. I also disagree that it automatically trumps what you consider to be a "weaker" source. Rexxs I understand and it makes sense, but when the results of reviews say the opposite I find it better to list them both - "A 2012 review found… however the 2013 review of reviews found….". It tells the actual story rather than making too bold of a statement.Herbxue (talk) 18:42, 7 July 2014 (UTC)

Your changing your previous comments after I replied to your comment. Later I explained it was the paragraph ending with the reference 89. QuackGuru (talk) 22:18, 7 July 2014 (UTC)

The weaker Vickers source was criticised by two other sources. How is a systematic review of systematic reviews not a stronger source? QuackGuru (talk) 18:51, 7 July 2014 (UTC)

It is still a reliable source. I am reading the Ernst review of reviews and his findings are pretty mixed and nuanced. Where in the paper does he conclude that there is no difference between sham and verum? I do not see that in the source. He does report on reviews that found no difference for back pain, but he does not generalize that to "acupuncture is no better than placebo for pain". He is reporting on one review, not making a conclusion based on all of them Herbxue (talk) 20:00, 7 July 2014 (UTC)

The other sources may be reliable but they are still weaker. The review does not say it is no difference for reducing pain in general but it found real and sham was no better in reducing pain for acupuncture. The reference cited was [89] at the end of the paragraph. QuackGuru (talk) 20:40, 7 July 2014 (UTC)

But the conclusion to THAT source says "It is concluded from this systematic review that acupuncture combined with exercise is effective for shoulder pain after stroke. It is recommended that future trials be carefully conducted on this topic." Ernst is commenting specifically on one review for back pain.Herbxue (talk) 21:16, 7 July 2014 (UTC)

That is the wrong source you cited (PMID 22924414). We are not using the [89] reference (PMID 19174438) cited by the 2011 source. This is the correct source (PMID 21440191). QuackGuru (talk) 21:30, 7 July 2014 (UTC)

OK so you linked to an abstract of Ernst 2011, but I think you meant ref 89 in that paper which is a 2009 systematic review ([107].) That looks like a reliable source but it does not justify say a review of reviews finds… and then giving the conclusion of just one of those reviews. Again, Ernst does not generalize on the question of acu vs sham in the 2011 source. Herbxue (talk) 21:40, 7 July 2014 (UTC)

We use the 2011 source not the older source cited by the 2011 source. The 2011 source does review for acupuncture in reducing pain real acupuncture is no better than sham. QuackGuru (talk) 21:46, 7 July 2014 (UTC)

Are you using this text from the 2011 source?

"This view was further strengthened by a recent randomized controlled trial in patients with osteoarthritis examining the effects of acupuncturists’ communication style [128].Real and sham acupuncture were both more effective in reducing pain than no acupuncture at all, but real acupuncture was no better than sham. Moreover, a communication style generating high expectations in patients resulted in improved outcomes compared to a normal style, regardless of the type of acupuncture administered. In the primary studies included in the systematic reviews evaluated above, the risk of bias was often considerable. Adequately controlling for nonspecific effects in future is likely to demonstrate that acupuncture has no or few specific effects on pain [89]."

That is Ernst telling us about the findings of a primary source, not Ernst's conclusions based on his review of all the reviews.Herbxue (talk) 22:46, 7 July 2014 (UTC)

No, that is Ernst et al drawing conclusions from their analysis of the sources they studied. It is precisely for the analysis performed by published experts in the field that we value secondary sources so highly. What we don't value is amateur non-experts trying to carry out their own analysis of sources. From WP:MEDASSESS:

• "Assessing evidence quality" means that editors should determine the quality of the type of study. Editors should not perform a detailed academic peer review. Do not reject a high-quality type of study due to personal objections to the study's inclusion criteria, references, funding sources, or conclusions.

What the authors chose to comment on is significant. They have studied all of the sources they reference and have drawn their conclusions from that study. You're not qualified to decide that a given conclusion in a secondary source isn't based on the review they have undertaken, and the sooner you understand that, the sooner we can move forward. --RexxS (talk) 13:01, 8 July 2014 (UTC)

Right, which is why we shouldn't cite Ernst '11 for sham = true since that's not the conclusion of his review of reviews, but rather a comment he makes about a subset of the reviews he covers. We'd cite the individual reviews for the sham = true statement, but weighting appropriately given other sources that don't say that. --Middle 8 (leave me alone • talk to me • COI?) 13:10, 8 July 2014 (UTC)

Of course it's the authors' conclusions and you have no way of knowing what factors they weighed when they wrote a conclusion like "real acupuncture was no better than sham". The context for that conclusion is set by the preceding part of the paragraph where we find it - wherein the authors chose to highlight: (1) an RCT showing sham acupuncture is as good as true acupuncture for alleviating chronic low back pain (and both are better than usual care); and (2) an RCT on osteoarthritis that strengthened that view. The fact is that the statement "Real and sham acupuncture were both more effective in reducing pain than no acupuncture at all, but real acupuncture was no better than sham." is accurate in context and is a clear expression of the authors' conclusions on the issue. The whole point of secondary sources is that they analyse the literature and the authors come to conclusions on the sources they have examined. If they draw the attention of the reader to a particular source, they are indicating its significance, and you are mistaken to suggest you can drill down to the sources a review is based upon. Ernst et al are qualified to make their analyses of those source: you and I are not. --RexxS (talk) 13:45, 8 July 2014 (UTC)

Another convenience break

Per BRD, I'm reverting QuackGuru's series of bold edits to the lede that made it say A systematic review of systematic reviews found that for reducing pain real acupuncture was no better than sham.[5]". (I will restore his other minor changes to the body.) That edit is an inaccurate reading of the source, Ernst '11, who in fact says

• "Cherkin et al. [14] have shown that, for chronic low back pain, individualized acupuncture is not better in reducing symptoms than formula acupuncture or sham acupuncture with a toothpick that does not penetrate the skin."

That's Ernst's summary of one review trial, Cherkin, and he says similar things about another review trial (Suarez-Almazor [128]) and considers these results very important, but it is certainly not the conclusion of Ernst '11 itself. He also says:

• "the majority of the early reviews [which were reviewed in an earlier overview similar to the present one] arrived at negative conclusions [85], while the majority of the 57 recent reviews were positive. Yet there are many contradictions and doubts..."

IOW, the majority of recent review do find true acu more effective than sham. Ernst is dubious about this finding, and thinks it will likely change, but that's his prediction and not the present situation. So, the source doesn't support the statement for which it's cited. This source-citation disparity repeats the same error we had with Moffet, which did not support the statement "There is no difference between "true" acupuncture and "sham" acupuncture (cf. top of subsection; diff). In both cases the source was cited for a general statement when in fact it was commenting on just a subset of the literature.
Also a general suggestion, I think we should stop editing the lede so aggressively (other than pruning cruft and minutiae). Seems like a poor use of editorial energy while the body needs so much work. --Middle 8 (leave me alone • talk to me • COI?) 13:02, 8 July 2014 (UTC) edited 15:26, 8 July 2014 (UTC)

I'm restoring the version prior as the text was properly sourced. Above you can see I've rejected your assertion that the source doesn't support the statement.

• "A systematic review of systematic reviews found that for reducing pain real acupuncture was no better than sham." (the text you removed)

is definitely supported by:

• "Real and sham acupuncture were both more effective in reducing pain than no acupuncture at all, but real acupuncture was no better than sham." (the source you discard)

Your assertion in the edit summary that "source comments on subset of literature reviewed" is nonsense. You're not entitled to do amateur peer-reviews of secondary sources by policy. All secondaries may choose to comment on whatever literature they decide, and you don't get to discard the review's conclusions on the grounds you don't like the ones they highlight.

As for your "IOW the majority of recent review do find true acu more effective than sham", you're drawing that conclusion out of thin air. Anyone reading the review can see that its text introducing your quote relate to "reviews of acupuncture for pain" - 17 earlier and 57 more recent. Immediately after your cherry-picked quote, the review goes on to say "For instance, there is no plausible reason why acupuncture should reduce pain in some conditions while failing to work in many others.", making it clear that the authors are discussion the effectiveness of acupuncture in reducing pain, not the difference between "true" and "sham" acupuncture. You're misusing this source to fabricate a conclusion that it does not state, and furthermore using that falsification to contradict the conclusion that it unambiguously does make: "Real and sham acupuncture were both more effective in reducing pain than no acupuncture at all, but real acupuncture was no better than sham". --RexxS (talk) 14:16, 8 July 2014 (UTC)

Rexxs did you READ the source? That is not a conclusion - the authors clearly are referring to one PRIMARY study, NOT EVEN A REVIEW! and then speculating that there will be more like it. As far as conclusions go there is the reviews support only few conditions, with many contradictions, and that better training should reduce adverse events. Those are the conclusions of Ernst 2011. Seriously, read it one more time before you decide this is worth fighting over.Herbxue (talk) 14:32, 8 July 2014 (UTC)

Yes I read it and I'm absolutely sure that when the review discusses several of earlier sources and then finally states:

• "Real and sham acupuncture were both more effective in reducing pain than no acupuncture at all, but real acupuncture was no better than sham."

they are reaching an unambiguous conclusion. It's a significant conclusion of a reliable secondary source and when you remove properly sourced text in the way you have, you are editing disruptively to suppress a view that you do not agree with. Please revert yourself, or I'll seek sanctions against you for this disruption. --RexxS (talk) 14:59, 8 July 2014 (UTC)

(e/c, but this applies to your comment at 14:59 too) Hi RexxS, please see your talk page... benefit of the doubt, please, re "cherry-picking". Please, cool heads! Did you read this comment? That was my way of saying, "please, let's make sure we're all familiar with the topic area before plunging too far ahead". Here's some useful background. When you say, "making it clear that the authors are discussion the effectiveness of acupuncture in reducing pain, not the difference between "true" and "sham" acupuncture," that tells me that you think effectiveness and the difference between "true" and "sham" are separate things. They're not. And then what Herbxue says is completely correct. "Real and sham acupuncture were both more effective in reducing pain than no acupuncture at all, but real acupuncture was no better than sham." is preceded by a whole paragraph that makes it clear he's talking about a particular trial (which is what I mean by "subset", and a small one at that being one paper, albeit a high-quality one). And what I said about the majority of recent reviews is intended to be about those for pain -- I'm talking about Ernst's statement "the majority of the 57 recent reviews were positive". Positive meaning efficacy. Meaning true > sham. That's the message of the pain aspect of the paper, combined with his prediction that in the future these results are likely to change. Please take some time to get up to speed on these basic aspects of the topic area; it will save all of us wikistress! --Middle 8 (leave me alone • talk to me • COI?) 15:24, 8 July 2014 (UTC) edited 15:28, 8 July 2014 (UTC)

Please don't patronise me. I'm more than familiar with WP:CPUSH. Your basic premise that 'positive efficacy' is equivalent to 'true > sham' is completely fallacious as I've patiently explained to you. Evidence of effectiveness does not imply any conclusive effect because of the placebo effect. True and sham could be utterly equivalent as long as both had the same placebo effect (which would then be pretty likely, of course). --RexxS (talk) 17:31, 8 July 2014 (UTC)

Sigh... see [108]. --Middle 8 (leave me alone • talk to me • COI?) 18:13, 8 July 2014 (UTC)

I'm gonna paste in the relevant paragraph (one para = fair use, I think) so editors can see for themselves what the context is for this sentence: "Real and sham acupuncture were both more effective in reducing pain than no acupuncture at all, but real acupuncture was no better than sham." Clearly this is not a ringing endorsement of acu, but I'm not saying it is, I'm just talkin' context:

“ These findings should be seen in the light of recent results from high-quality randomized controlled trials. Cherkin et al. [14] have shown that, for chronic low back pain, individualized acupuncture is not better in reducing symptoms than formula acupuncture or sham acupuncture with a toothpick that does not penetrate the skin. All 3 forms of acupuncture, however, were more effective than usual care. The authors consider, therefore, that the benefits of acupuncture ‘‘resulted from nonspecific effects such as therapist conviction, patient enthusiasm, or receiving a treatment believed to be helpful’’ [14]. This view was further strengthened by a recent randomized controlled trial in patients with osteoarthritis examining the effects of acupuncturists’ communication style [128]. Real and sham acupuncture were both more effective in reducing pain than no acupuncture at all, but real acupuncture was no better than sham. Moreover, a communication style generating high expectations in patients resulted in improved outcomes compared to a normal style,

regardless of the type of acupuncture administered. In the primary studies included in the systematic reviews evaluated above, the risk of bias was often considerable. Adequately controlling for nonspecific effects in future is likely to demonstrate that acupuncture has no or few specific effects on pain [89].[109]

”

happy editing, --Middle 8 (leave me alone • talk to me • COI?) 16:38, 8 July 2014 (UTC)

But the effectiveness of acupuncture and the difference between "true" and "sham" are separate things. If both were to be effective placebos, then both could be equally effective, and there would be no difference. The presence or absence of effectiveness is not therefore a determinant of any difference between between true and sham acupuncture. The review is entitled "Acupuncture: Does it alleviate pain and are there serious risks?"; You simply cannot draw your own unsupported conclusion that when Ernst et al start the discussion section they are addressing the difference between sham and true. That is not even mentioned until much later in the section.

What Herbxue says is a long way from correct. If you remember, the "Discussion" section began by surveying the types of pain for which they had reviews - commenting that only treatments of low back pain and osteoarthritis had a majority of positive high quality reviews. The paragraph that contains the text in question starts by mentioning two high-quality RCTs: one that showed the lack of difference between true and sham on chronic back pain; the other supporting that same conclusion for osteoarthritis - thus counter-balancing the evidence of effectiveness in those two areas with the point that the effectiveness exists for both true and sham. It is after that Ernst et al state "Real and sham acupuncture were both more effective in reducing pain than no acupuncture at all, but real acupuncture was no better than sham". That is not the review authors quoting other studies, but their own conclusion, not referenced to any particular source, but a statement of what they had found in reviewing the current literature. You will see that they chose to balance high-quality studies against earlier reviews; that is their prerogative - even if it is not ours - because they are the published experts in the field, and we are merely amateur editors who have to agree to follow what reliable secondary sources say, not place our own interpretations on them. --RexxS (talk) 17:17, 8 July 2014 (UTC)

RexxS, to be honest, this is getting into WP:CHEESE territory. Just as "significant" is shorthand for "statistically significant", "efficacy" (or effectiveness) is shorthand for "efficacy beyond placebo". It is understood that people respond to placebos. But I guarantee you that when Ernst talks about efficacy, he means what I just said. What is the placebo? Sham acu. What is efficacy beyond placebo? Greater response to true acu than to sham acu. (What is disruptive? This discussion.) --Middle 8 (leave me alone • talk to me • COI?) 18:10, 8 July 2014 (UTC)

I'm glad you posted this at Project Med, we definitely need more eyes on this paper.Herbxue (talk) 17:43, 8 July 2014 (UTC)

Thank you. When it's just a few of us debating we can get entrenched in our own positions, right or wrong. The folks at WPMED are very experienced, so they can almost always break the log-jam on the talk page, if we can just get them to spend the time on helping with our petty squabbles. --RexxS (talk) 18:00, 8 July 2014 (UTC)

We should do that earlier next time. --Middle 8 (leave me alone • talk to me • COI?) 19:41, 8 July 2014 (UTC)

According to this comment by Middle 8 it seems he knows the text is indeed sourced. He didn't like the sourced text because he claims it was "not the overall findings of his review of reviews"? QuackGuru (talk) 19:47, 8 July 2014 (UTC)

Yet another convenience break

We are talking about the paragraph "These findings should be seen in the light of recent results from high-quality randomized controlled trials..." in Ernst et al (2011), right? There are three citations in that paragraph: two randomized controlled trials, and a systematic review of randomized controlled trials. All trials include sham treatments. It is difficult to accept the "particular trial" justification for deleting the sourced content. The fact that Ernst et al provided a specific example does not imply that their conclusions are limited to that example. Indeed there is a second example, plus a systematic review. vzaak 21:04, 8 July 2014 (UTC)

@Vzaak: -- did you read the original? email me if you need a copy. Re your comments: [110]. --Middle 8 (leave me alone • talk to me • COI?) 00:04, 9 July 2014 (UTC)

Yes that is the passage. Show me where he makes a conclusion though - the sentence is clearly referring to the clinical trial he is discussing, using past tense. Not a conclusion, which would be worded something like "Therefore, it can be assumed that acupuncture and placebo are equally effective" (present tense). Also, if it were a conclusion of his review, why is it not in the abstract, or at the end of the paper? Herbxue (talk) 22:38, 8 July 2014 (UTC)

Correction. You did think the sentence is referring to a 2009 review. QuackGuru (talk) 22:44, 8 July 2014 (UTC)

What difference does that make? I was trying to understand how you justified your over-generalized text.Herbxue (talk) 22:48, 8 July 2014 (UTC)

It is not a typical paper. It is a systematic review of systematic reviews. User:Herbxue, I think you should revert your edit. Your edit summary claims "Was not a conclusion, was reference to a primary source. Per talk." But it was in reference to a 2009 review according to your previous comment. The [89] reference (PMID 19174438) was a 2009 review cited by the systematic review of systematic reviews. The consensus is against you on this. See Wikipedia talk:WikiProject Medicine#Acupuncture again. QuackGuru (talk) 23:05, 8 July 2014 (UTC)

Exactly, it's a review of reviews, and those RCT's cited on p.762 aren't reviews, and not part of the findings. Rather they're used as part of discussion of findings. Which are on p.761. This is a pretty amusing flirtation with Wikiality... I kinda hope consensus does get it wrong, so I can email Ernst and hopefully give him a good laugh. --Middle 8 (leave me alone • talk to me • COI?) 00:10, 9 July 2014 (UTC)

• Herbxue, the paragraph in toto relates to sham treatments. All the sources in the paragraph relate to sham treatments. Your objection seems more to do with the particular wording in the Wikipedia article. I would simply write, "At least two recent high-quality randomized controlled trials found that for reducing pain real acupuncture was no better than sham", full stop, with citation to Ernst et al (2011). No need for "A systematic review of systematic reviews...", which seems to be the locus of your dispute. vzaak 00:43, 9 July 2014 (UTC)

Vzaak, per MEDRS we cite reviews not RCT's, and systematic reviews (SR's) are better than not, and an SR of SR's like Ernst is of very high quality (though not unsurpassingly so). So, we like to cite directly to that. --Middle 8 (leave me alone • talk to me • COI?) 01:44, 9 July 2014 (UTC)

I wouldn't cite the RCTs. As I said, I would cite the Ernst paper. We are merely referencing the synthesis that Ernst did, not making our own synthesis. Ernst et al singled out those papers, and so may we. The cite-conclusions-only rule applies to primary sources (if they are even considered in the first place). For reviews (and reviews of reviews), we may cite from the main text. vzaak 02:33, 9 July 2014 (UTC)

See my CPUSH answer opposing those RCT's at WT:MED [111] note: some snark may be present --Middle 8 (leave me alone • talk to me • COI?) 06:27, 9 July 2014 (UTC) To clarify, said "snark" refers to my sarcasticaly calling my own comments "CPUSH". What I said at WT:MED is not snark. --Middle 8 (leave me alone • talk to me • COI?) 00:36, 10 July 2014 (UTC)

Stating "At least two recent high-quality randomized controlled trials" is not accurate. It was Ernst et al (2011) who found that for reducing pain real acupuncture was no better than sham. QuackGuru (talk) 00:51, 9 July 2014 (UTC)

That's a tendentious reading of the source (a fact that I think should be obvious to any editor who is generally good at carefully reading sources), and in the current environment I'm not sure I'd push it. --Middle 8 (leave me alone • talk to me • COI?) 01:50, 9 July 2014 (UTC)

• In my view, QuackGuru seems to have a fine interpretation of Ernst's position in his 2011 paper. But in the interest of not getting caught up over specific wording about study design, I have made a couple of edits that I think capture the spirit of the literature that we're looking at.[112] Thoughts? NW (Talk) 02:34, 9 July 2014 (UTC)

You've read the full text of Ernst '11? For SAE's he (correctly) uses wording like "SAE's after acupuncture", not "attributable to". Five deaths in a decade known in the world = weighty enough in the lede? Pretty sure we don't do this for lots of other topics where the occasional tragic accident happens. Is there a frequency threshold below which we shouldn't? Genuine question. --Middle 8 (leave me alone • talk to me • COI?) 06:27, 9 July 2014 (UTC)

PMID 21440191, right? Ernst does evaluate the causality between acupuncture and SAE in the reported case reports (or perhaps the case reports themselves assess the causality; I didn't look into that part much). In either case, they are largely "certain", not "possible" or even "likely". With that said, I largely agree with your stance. I think we can generalize about SAEs but should avoid going into specifics, as Table 3, a collection of newspaper articles or case studies from practitioners who took the not-insubstantial amount of time to write up and submit the report to a journal, could understate things by anywhere from zero to four orders of magnitude. Unfortunately, as there is no systematic surveillance as there is for prescription drugs (see Ernst E, Pain 152(9): 2184–6), that becomes just too large of a range to say anything useful about in the lead. I just didn't want to be the one to unilaterally alter the meaning of that paragraph, but while it is important to take reported safety figures into consideration in the lead, I think it is also important to mention the range of uncertainty. NW (Talk) 12:57, 9 July 2014 (UTC)

Agree there's a range of uncertainty. Four OOM's? Is that sourced? Anyway going back on wikibreak, see you all later. --Middle 8 (leave me alone • talk to me • COI?) 07:16, 11 July 2014 (UTC)

No, that's an educated guess for an absolute upper limit. NW (Talk) 13:27, 11 July 2014 (UTC)

Bold edit to resolve the issue

Ernst et al are referring to the recent RCTs, not making a broad sweeping conclusion that incorporates all studies. In the Ernst paper, the following text is obviously referring to Suarez-Almazor et al[128]: "Real and sham acupuncture were both more effective in reducing pain than no acupuncture at all, but real acupuncture was no better than sham. Moreover, a communication style generating high expectations in patients resulted..." (my bolding). Note the title of [128]: "A randomized controlled trial of acupuncture for osteoarthritis of the knee: effects of patient-provider communication". More explanation in my comment at WP:MED.[113]

However this is no excuse to delete the material from the article. I have made this bold edit which is the closest I can imagine to satisfying (or at least satisficing) all parties: "A systematic review of systematic reviews highlighted recent high-quality randomized controlled trials which found that for reducing pain real acupuncture was no better than sham acupuncture."[114] vzaak 18:46, 9 July 2014 (UTC)

I am happy with that text. Axl ¤ [Talk] 19:35, 9 July 2014 (UTC)

It is a good compromise with the in-text attribution. QuackGuru (talk) 19:39, 9 July 2014 (UTC)

Agreed. Herbxue (talk) 19:46, 9 July 2014 (UTC)

Nicely done. At some point I think we'll want to specify that in addition to those RCT's a lot of sec sources find the same thing (sham = true), as well as indicating there are a few that don't, like Vickers, quoted above. I'm sorry, Vzaak, if my reluctance to cite these RCT's seemed tendentious. We talk a lot about efficacy in the lede, and saying "sham = true" is the exact same thing as saying "no efficacy", which a lot of sources that we cite find to be the case; therefore, I was (and remain) unsure about why sham needs to be in the lede at all, but this compromise is fine. If I really were being tendentious I wouldn't suggest using even better MEDRS's to reference sham (e.g. the sec sources you can find in Colquhoun & Novella editorial). --Middle 8 (leave me alone • talk to me • COI?) 07:15, 11 July 2014 (UTC)

Cost-effectiveness in lede: why?

This is way too granular for the lede: A meta-analysis found that acupuncture for chronic low back pain was cost-effective as a complement to standard care, but not as a substitute for standard care except in cases where comorbid depression presented[15] while a systematic review found insufficient evidence for the cost-effectiveness of acupuncture in the treatment of chronic low back pain.[16] I don't think we need any of that since there's nothing especially interesting, plus or minus or sideways, about cost-effectiveness. --Middle 8 (leave me alone • talk to me • COI?) 11:40, 8 July 2014 (UTC)

Per WP:LEDE, we should summarise the cost-effectiveness section. QuackGuru (talk) 17:25, 8 July 2014 (UTC)

But could we adequately summarise it without quite so much detail? I'd be happy to discuss any proposed alternate wording. --RexxS (talk) 17:34, 8 July 2014 (UTC)

Middle 8 is not proposing to tweak the text. I think he wants to delete both sentences. I open to suggestions to improving the text if needed but we can't delete a summary. QuackGuru (talk) 17:39, 8 July 2014 (UTC)

I hope it goes without saying that anything I propose here is open to compromise... since this is Wikipedia, last time I checked. Cut it or prune it, whatever, just a suggestion. --Middle 8 (leave me alone • talk to me • COI?) 19:40, 8 July 2014 (UTC)

Per LEDE, you should summarize the article. In an article with seven ==major sections== and twenty six (26) ===subsections===, there is no rule that says you have to mention every single subsection, much less to give two sentences to one of them—especially since there are only 23 sentences in total. WhatamIdoing (talk) 00:07, 9 July 2014 (UTC)

Menopause syndroms

• Frisk, Jessica W.; Hammar, Mats L.; Ingvar, Martin; Spetz Holm, Anna-Clara E. (30 January 2014). "How long do the effects of acupuncture on hot flashes persist in cancer patients?". Supportive Care in Cancer. 22 (5): 1409–1415. doi:10.1007/s00520-014-2126-2.

• Chiu, Hsiao-Yean; Pan, Chieh-Hsin; Shyu, Yuh-Kae; Han, Bor-Cheng; Tsai, Pei-Shan (7 July 2014). "Effects of acupuncture on menopause-related symptoms and quality of life in women on natural menopause: a meta-analysis of randomized controlled trials". Menopause (journal): 1. doi:10.1097/GME.0000000000000260.

• Dodin, Sylvie; Blanchet, Claudine; Marc, Isabelle; Ernst, Edzard; Wu, Taixiang; Vaillancourt, Caroline; Paquette, Joalee; Maunsell, Elizabeth; Dodin, Sylvie (2013). "Acupuncture for menopausal hot flushes". doi:10.1002/14651858.CD007410.pub2. PMID 23897589. {{cite journal}}: Cite journal requires |journal= (help)
• Cho, S. H.; Whang, W. W. (2009). "Acupuncture for vasomotor menopausal symptoms". Menopause. 16 (5): 1065–1073. doi:10.1097/gme.0b013e3181a48abd. PMID 19424092.
• There is currently MEDRS compliant sources in Acupuncture#Other conditions. One is a 2013 Cochrane review and the other is a 2009 systematic review. QuackGuru (talk) 04:41, 11 July 2014 (UTC)

Commenting on the Chiu paper today, Steven Novella called the results "entirely negative" and concluded ..."A reasonable person can only conclude that acupuncture does not work, and that all the clinical research consistently shows that acupuncture conveys only illusory and nonspecific placebo effects for subjective symptoms." [115]

Violation of WP:LEDE and WP:V

This change to the first sentence is a clear violation of WP:V and WP:NPOV policy.

"Acupuncture therapy is believed to have developed in China over thousands of years and refers to the stimulation of precisely defined, specific points on meridians (or channels) that lie along the surface of the body and within organs. Stimulation of acupoints can be accomplished through a variety of methods including application of heat, pressure, or laser or insertion of thin needles.1"[116] The source does not support the change. QuackGuru (talk) 19:13, 13 July 2014 (UTC)

More original research added to the lede using a 2005 primary source. QuackGuru (talk) 20:08, 13 July 2014 (UTC)

I don't see why acupressure should be the second sentence in the lede. It's way off topic. I removed the sentence and the ref. Perhaps an appropriate place in the body. Jim1138 (talk) 19:12, 14 July 2014 (UTC)

It is in the body under Acupuncture#Related practices using a systematic review. I have restored the other text that was deleted. QuackGuru (talk) 19:42, 14 July 2014 (UTC)

The restored text is appropriate. It is a feature of acupuncture practice that needles are not the only tool used to stimulate the points or channels.Herbxue (talk) 19:56, 14 July 2014 (UTC)

So that there is no "puncture" ? isn't that then something else? Alexbrn ^{talk|contribs|COI} 20:06, 14 July 2014 (UTC)

Well, depends on how we are defining acupuncture. In China it is referred to as 针灸 zhen jiu, literally "needles and moxibustion", and it is assumed that needling, moxa, cupping, gua sha, and other auxiliary techniques are performed by the acupuncturist.Herbxue (talk) 20:16, 14 July 2014 (UTC)

Sounds like "针灸" connotes a broader concept than English "acupuncture". In English the word acupuncture means "A method of medical treatment, originating in China, in which fine needles are inserted into the skin at specific points on the body surface" (OED primary definition). Alexbrn ^{talk|contribs|COI} 20:23, 14 July 2014 (UTC)

It would be misleading to claim that acupuncture uses only needles. Yes, I read up on acupuncture before I originally added the first sentence to the WP:LEDE. I also added the definition of traditional acupuncture to the lede. QuackGuru (talk) 20:43, 14 July 2014 (UTC)

The NHS says: "Acupuncture is a form of ancient Chinese medicine in which fine needles are inserted into the skin at certain points on the body". Alexbrn ^{talk|contribs|COI} 20:48, 14 July 2014 (UTC)

It is also true that in the US, in most states the practice act defines the practice as needle insertion, though many states also include the auxiliary techniques. If the article were just about the specific procedure of needle stimulation then the narrow definition would fit, but the article has a broader scope. "Acupuncturists" are almost always trained in the auxiliary techniques, with moxa being considered by many as a primary technique with equal importance to needle insertion.Herbxue (talk) 20:58, 14 July 2014 (UTC)

(Add) And Ernst defines it as an "Insertion of a needle into the skin ...". We have a separate article on moxibustion. Alexbrn ^{talk|contribs|COI} 21:00, 14 July 2014 (UTC)

Indeed! While acupuncturists might normally do more than strict acupuncture, we have separate articles here. I don't expect Herbxue to reply, since he is now topic banned for six months. -- Brangifer (talk) 03:01, 15 July 2014 (UTC)

We have RS that acupuncture involves far more than needle insertion. It does not benefit the reader to alter the broader definition of acupuncture. QuackGuru (talk) 04:56, 15 July 2014 (UTC)

My dear QG, that's from possibly the fringey-est journal in the world. Wouldn't you say the NHS, the OED and Ernst are better? Alexbrn ^{talk|contribs|COI} 05:38, 15 July 2014 (UTC)

It is common knowledge that acupuncture involves more than needles. Traditional acupuncture also involves more than needles. These are mundane claims. The current wording is correct and accurate. Shortening the definition would make the text factually incorrect. QuackGuru (talk) 05:47, 15 July 2014 (UTC)

It's not common knowledge. Sure, "what acupuncturists do" might include other things – but that's not the topic of this article. According to the strongest RS, acupuncture is needling into certain points on the body; a fringe journal defines "traditional" acupuncture (whatever that is) more broadly, citing a single Chinese source tellingly entitled "Acupuncture and moxibustion" (my emphasis). We need to follow the best sources. Alexbrn ^{talk|contribs|COI} 05:56, 15 July 2014 (UTC)

(add) Consider also:

• Karchmer EI (2009). "Acupuncture and Moxibustion". In Davis EK (ed.). Taylor & Francis. p. 4. ISBN 978-0-415-77716-2 http://books.google.com/books?id=2rLBvrlKI7QC&pg=PA4. Acupuncture refers to the therapeutic manipulation of needles at specific sites on the body, and moxibustion to the therapeutic warming or cauterizing of those sites with slow-burning moxa leaves. They are commonly used together in clinical practice ... {{cite book}}: |work= ignored (help); Missing or empty |title= (help)

The English definition for the word acupuncture is not factually accurate.

We are using good RS to verify the first sentence for the definition of acupuncture. The second sentence is the explanation for "traditional acupuncture". Both statements are accurate and from a NPOV. QuackGuru (talk) 06:12, 15 July 2014 (UTC)

Sourced text was deleted

And some original research was added. At least three sources are obviously MEDRS violations. QuackGuru (talk) 17:07, 17 July 2014 (UTC)

Can you please specify the MEDRS violations? Jayaguru-Shishya (talk) 17:49, 17 July 2014 (UTC)

I removed the primary sources and poor sources. QuackGuru (talk) 22:16, 17 July 2014 (UTC)

Can you please specify the MEDRS violations, what were the primary sources and what were the poor sources you removed? Thanks. Jayaguru-Shishya (talk) 22:27, 17 July 2014 (UTC)

Please check the edit history and each edit summary. QuackGuru (talk) 22:31, 17 July 2014 (UTC)

Doesn't answer my question. Can you please provide a description of all your edits you made on the article? Thanks. Jayaguru-Shishya (talk) 22:36, 17 July 2014 (UTC)

Impact factor

Hi! I was wondering, what is the minimum impact factor of a source according to WP policies to be sufficient enough in order to be published in an article? I couldn't find such information myself, but I think you can help! Thanks! Jayaguru-Shishya (talk) 22:34, 17 July 2014 (UTC)