Talk:Homeopathy/Archive 12

From Wikipedia, the free encyclopedia

 This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
← Archive 11 Archive 12 Archive 13 →

Contents

Homy outline

New outline on homeopathy and classical homeopathy Please add your editions in these versions --Homy 18:42, 7 August 2006 (UTC)

Academic Dishonesty

Whoever added in these references as part of the criticism of the 2005 Lancet meta-analysis, you should be ashamed of yourself. Did you even read them?

  • [1] Volume 11, Number 5, 2005, pp. 759–761; THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE

Mary Ann Liebert, Inc. A “Maddox Effect”? A Reason to Adopt Time Series Protocols in Tests of Homeopathic Remedies ALEX HANKEY, Ph.D.

This article doesn't even MENTION the Lancet study, and that's still better than the second one:

I would love to hear someone (maybe the person who added it?) explain why an article from 1998 is being used to directly criticize a study from 2005. Did homeopaths invent time travel when I wasn't looking? Or are homopaths so bloody desperate for any scientific voice that they need to deliberately confuse a meta-analysis done in 1997 with the one done in 2005? I know, I know, never attribute to malice what can be attributed to mere ignorance.

There's no question that I'm removing these references, and I'm also adjusting the text to reflect the due weight that the remaining online, effectively self-published, criticisms should be given. I further recommend that in the absence of any other criticisms that meet basic standards for scientific discourse (i.e. published in a peer-reviewed journal), this section should probably be removed. T.J.C. 03:21, 14 September 2006 (UTC)

GA review

As an offshot of the GA discussion, I am attempting at a quickie "review" of the article. I do not care who put where and why, I am just reviewing the article in the version that appeared to me on 03:40, 14 September 2006 (UTC). Some more detailed stuff that occurred to me while reading:

  • summary section seems very OK to me, for what I know about the subject (that ain't much, to be sure)
  • the "Medical Similars" section is quite long and not divided into subsections, but rather with many short paragraphs. It is not the best example of structuring, and doesn't read too well either.
    • Comment Could go to classical homeopathy --Homy 07:24, 22 September 2006 (UTC)
Section divided into sub-sections for clarity.TimVickers 04:58, 14 September 2006 (UTC)
  • "Theory of disease" - is "germ theory" a widely used term, or just coined for the article?
This term dates back to the beginning of microbiology. Added Wilkilink. TimVickers 04:44, 14 September 2006 (UTC)
    • Comment Agree? --Homy 07:24, 22 September 2006 (UTC)
  • Some dates in the abovementioned section, and other places in the article, are put in square brackets. I believe that the WP standard is to use round brackets -> ( )
Square brackets replaced. TimVickers 04:43, 14 September 2006 (UTC)
  • Suddenly out of nowhere, James Tyler Kent pops out. If he really was as important as Heinemann in the development of homeopathy, he should be given the same amount of attention. It seems that the history of the homeopathic theory of disease ends up somewhere in the first two decades of the 19th century - this hardly makes for a comprehensive section, as I infer from the article that there were some changes to it afterwards
    • Comment Kent is not as important, but his repertory was. Vithoulkas brought homeopathy in 20th century but moderator User:Geni and user Jefffire want to keep homeopathy in the 19th as my additions about quantum physics and relativity prove.
  • "Homeopathy around the world" is de facto "Homeopathy in selected countries", and should use lower-level subheadings rather than just bolding of the country's name. It also reads a bit like a hodgepodge of information that didn't fit anywhere else. US, and to some extent UK, sections give a more or less complete overview of the history of homeopathy in those countries and the current relation to organized healthcare systems, while the other are merely one-paragraph notices varying in content.
  • "Classical versus non-classical homeopathy" - unreferenced, a bit confusing, and with weasel phrases.
  • "Number of homeopaths" - this section deals with UK exclusively, and the last paragraph is unreferenced and looks like original research.
Agreed, removed. (edit, since Peter doesn't seem to get it: Wikipedia articles should not be independently drawing these kinds of conclusions) T.J.C. 05:07, 14 September 2006 (UTC)
I have re-written this a bit to show this is an interpretation of the data presented in the paragraph above, not the presentation of novel data. TimVickers 15:20, 14 September 2006 (UTC)
Unfortunately, Tim, I would argue that any novel interpretation of the data that does not directly follow (i.e. that demand for homeopathy is growing because the number of homeopaths are growing) in order to make a claim qualifies as original research. See [2]. I will, hoever, abstain from further edits - hopefully others will weigh in on their opinions.
I agree, which is why I say these data suggest this interpretation. I do not say they prove this interpretation. TimVickers 15:42, 14 September 2006 (UTC)
  • Oh, and now the Flexner Report and Vtihoulkas pop out of nowhere, and those seem to be fairly important. This article is a mess!
    • Comment Vithoulkas is. Flexner Report doesn't ring a bell. --Homy 07:30, 22 September 2006 (UTC)
This report is discussed fully in the section above this Historical Note, but without mentioning its name, wikilink moved upwards to the main section and this fragment removed. TimVickers 22:11, 14 September 2006 (UTC)
  • The last paragraph of "Mechanism of action of homeopathic preparations" includes an external link in lieu of a proper reference.
Reference formatted. TimVickers 04:48, 14 September 2006 (UTC)
  • "Lancet 2005 study" - the last sentence needs to explain who criticized the Lancet study. It also looks weird detached from the main article.
Replaced selfpublished with peer-reviewed crits. Jedermann 16:16, 14 September 2006 (UTC)
"trial quality which was significantly better in the homeopathic trials than the randomly chosen biomedical trials" doesn't seem to be on a peer-reviewed article. note that letters are usually not peer-reviewed. --calvinchong
Shang et al clearly stated that the homeopathy trials were significantly better quality than the randomly chosen comparators. This section needs a NPOV rewrite, as the metaanalysis was of 110 trials, and gave a positive result. Only the 8 in the subgroup analysis were negative, and the critics allege data dredging since this subgroup was not prespecified. Shange et al also refer in the Discussion section to another subgroup of 8 URTI trials that were incontrovertibly positive. In addition, the PEK committee said (in 2003) that Shang et al 'had gone too far' in their interpretation of the results.Jedermann 10:52, 18 September 2006 (UTC)
    • Comment Agree --Homy 07:30, 22 September 2006 (UTC)
  • "Medical organizations' attitudes towards homeopathy" - a bit of a laundry list, perhaps could be better orgaised with subheadings.

OK, I've made it to the end of the article! More general impressions:

  • In an article on such a contoversial topic, no paragraph should go unreferenced, as almost all contain statements that could be disputed and dismissed as original research unless referenced. Moreover, I trust in the good will of the editors, that they do not "add" to the material sourced from the books/websites/journals listed, but it would be good to make sure that no original research passed as referenced material.
  • The article is poorly organized, with the division of the contents between sections not clearly defined, some sections relatively underdeveloped compared to the others, some consisting of anecdotary information rather than general overview and the feeling I got is that this article might not be complete with regards to that (i.e. providing a good, balanced overview on all major aspects of the subject).
  • Style leaves a bit to be desired at times, and the division into paragraphs and sections is pretty random. It would be good to afford this article a major overhaul and perhaps partial rewrite so that it would be easier to read.

This was not a proper review, and some of my comments are vague, as I do not have enough time to go into too much detail, do not know enough about the topic to assess e.g. completeness of coverage or NPOV, and the GA standards are also currently being discussed, so I am not sure how to apply them, but I believe this nomination can easily be failed for either of the abovementioned reasons. I will not fail it for the fact that I am not able to provide a comprehensive review, but the article simply does not seem to be on the stage of development at which it would be ripe for GAN. I guess this is one of the hardest topics to work on, so I also want to say that I really do appreciate the efforts done until now. Regards, Bravada, talk - 03:40, 14 September 2006 (UTC)

Inconclusive?

Linde et al 1997: Findings: The combined odds ratio for the 89 studies entered into the main meta-analysis was 2.45 (95% CI 2.05, 2.93) in favour of homoeopathy. The odds ratio for the 26 good-quality studies was 1.66 (1.33, 2.08), and that corrected for publication bias was 1.78 (1.03, 3.10). Four studies on the effects of a single remedy on seasonal allergies had a pooled odds ratio for ocular symptoms at 4 weeks of 2.03 (1.51, 2.74). Five studies on postoperative ileus had a pooled mean effect-size-difference of -0.22 standard deviations (95% CI -0.36, -0.09) for flatus, and -0.1 8 SDs (-0.33, -0.03) for stool (both p lt 0.05). Interpretation: The results of our meta-analysis are not compatible with the hypothesis that the clinical effects of homoeopathy are completely due to placebo. However, we found insufficient evidence from these studies that homoeopathy is clearly efficacious for any single clinical condition. Further research on homoeopathy is warranted provided it is rigorous and systematic.

Why is this reference used to support the 'no different from placebo' POV, when the authors say, "The results of our meta-analysis are not compatible with the hypothesis that the clinical effects of homoeopathy are completely due to placebo". One difficulty with reviewing this article is the absence of Author-Date references, as preferred by WP. (Linde et al 1997) in the text would immediately identify a faulty (and POV) interpretation of the findings. Jedermann 11:14, 18 September 2006 (UTC)

I think it's used as a reference because some of the articles they reviewed showed no effect. Jefffire 11:28, 18 September 2006 (UTC)
If the statement is about "some" or a few trials, it would not be worth making. And it's a truism that better quality trials will show smaller effects - in all areas of healthcare, not just homeopathy. But it's supposed to be about the majority of trials. However, Linde et al 1997 said their results favoured homeopathy to the extent that "If the result of new trials were to show no difference between homeopathy and placebo, we would have to add 923 trials with no effect with 118 patients in each in order to balance the two." The reference does not support the POV. Jedermann 11:55, 18 September 2006 (UTC)
But some of the components do. Could we not use it as a secondary source? Linde et al are now rather out of date anyway as a primary source, with the Lancet 2005 study furfiling their recommendation of further studies with greater scientific standards. Jefffire 12:17, 18 September 2006 (UTC)
An absolute majority of the individual randomized trials meeting criteria for inclusion in each of the big 'global' systematic reviews (Kleijnen, Linde, Shang) favoured homeopathy over placebo, but where in this article is that acknowledged? If anything it says the opposite, i.e. 'most' trials were negative. But that is not true. Replicating Linde 1997, Shang 2005 still found a positive effect for all 110 trials. Here, only the subgroup analysis of 8 trials is mentioned as if that were the whole study. NB, systematic reviews and metaanalyses are only ever secondary sources: the component trials are primary. Jedermann 14:49, 18 September 2006 (UTC)

Wikipedia is build on a foundation of accepting authority. The Lancet is the most authoritative source we have, with had the most rigorous meta-analysis. Therefore we present its findings as such. WP:NPOV. However the discussion is now drifting away from the original points. Jefffire 15:49, 18 September 2006 (UTC)

Why have you reverted a selfcontradictory statement, which uses a positive Lancet 1997 metaanalysis to support the POV that many trials were negative, when the Lancet found they were mostly positive? How was consensus achieved to state the opposite of the Lancet conclusions? Likewise Lancet 2005 found overall positive results in 110 trials, but a subgroup analysis of 8 trials found negative results. Jedermann 17:03, 18 September 2006 (UTC)
I'm sorry, but you are not understanding my comments. The articles says that "The results of many medical trials show no difference between homeopathic treatments and placebos", and within the reference it show that many trials do show no difference between homeopathic treatments and placebos. The actual conclusion of the meta-analysis is not relevent in this case. If you wish to put in more references then go ahead. Jefffire 19:05, 18 September 2006 (UTC)
With the greatest respect for your NPOV, Jefffire, reference 14 does not support the interpretation put upon it here. If 'many' trials were no different from placebo, then the metaanalysis could not have reached a conclusion that favoured homeopathy, and was robust in the sensitivity analyses. That is not my POV, it is the result of the 1997 metaanalysis, published in the Lancet. If you want to demonstrate 'many' negative trials you will have to find a different reference. I don't know of one. Jedermann 22:04, 18 September 2006 (UTC)
These are refs 14-17 with my comments

^ Linde K, Clausius N, Ramirez G, Melchart D, Eitel F, Hedges LV, Jonas WB. Are the clinical effects of homeopathy placebo effects? A meta-analysis of placebo-controlled trials. Lancet. 1997 Sep 0;350(9081):834-43. Demonstrated robust superiority over placebo, even for better quality trials

^ Homeopathy & Meta analysis - hopeless, as it just repeats Linde 1997 and Kleijnen 1991 (which is discussed later anyway)

^ Shang A, Huwiler-Muntener K, Nartey L, Juni P, Dorig S, Sterne JA, Pewsner D, Egger M (2005). "Are the clinical effects of homoeopathy placebo effects? Comparative study of placebo-controlled trials of homoeopathy and allopathy". Lancet 366 (9487): 726-32. PMID 16125589. - Fine as it builds on Linde 1997, and claims that higher quality larger trials were equivalent to placebo.

^ Homeopathy versus Standard Treatments a summary of a paper by E. Ernst published in Perfusion 1999;12:13-15. Bandolier Journal. Hopeless - and irrelevant as not about placebo question, and the trials were so poor that nothing can be said either way.

I've been back over the attempted edits to these 2 sentences, and can see where the problem came in. My revision made sense of the conflicting data from 2 Lancet metaanalyses, and should offend noone. If you are not prepared to undo your revert I must ask for arbitration. Jedermann 09:03, 19 September 2006 (UTC)

Arbitration? Good grief, we are still discussing the matter, and no-one else has contributed yet, and the other paths have not been explored yet. Such an action would be grossly premature, disproportionate, and would not endear you to the admins.

Anyway, it is a simple matter to use anouther reference if you are unsatisfied with my rational for the current one, even though I have explained quite clearly why I correct in this. Jefffire 11:46, 19 September 2006 (UTC)


Jedermann, it is incorrect to say "the results of most clinical trials show a difference between homeopathic treatments and placebos" since that assumes that most clinical trials of homeopathy are methodologically sound enough to draw that conclusion. One source estimates the number of controlled trials where it is valid to make such a claim (i.e. controlled trials) to be 29%. (PMID 11801202)

The most accurate summary statements we can make regarding the scientific evidence for the efficacy of homeopathy are:

  • many homeopathic trials that may show a positive effect are unconvincing due to their poor quality
  • higher quality homeopathic trials are more likely to support the hypothesis that the apparent clinical effects of homeopathy are due to the placebo effect
  • the most recent and largest peer-reviewed meta-analysis of homeopathic trials provides no strong evidence to dismiss the hypothesis that the clinical effects of homeopathy are due to the placebo effect

I believe all three of these points are adequately defended by the 2005 Lancet review, with perhaps the exception of the first, which may be better defended by the article I cited here. If there is no objection to the above statements, I can try to rewrite the offending two lines to summarize the above points tonight. T.J.C. 20:07, 19 September 2006 (UTC)

Thanks, I look forward to reading it. Checking again the forest plot on pp.836-7 of Linde at al 1997, there are few trials with odds ratio between 0.1 and 1, not the 'many' which I object to. This confirmed Kleijnen 1991, where >60% trials favoured homeopathy. Sure, they may all be flawed, but that's a matter for posthoc interpretation. Can you remove the reference which did not look at placebo-controlled trials: ^Homeopathy versus Standard Treatments a summary of a paper by E. Ernst published in Perfusion 1999;12:13-15.
In general, for instance in the scientific evaluation section of the article (which is badly organized at present), it would be better to use high-quality "reviews of reviews" as the main source, so people can't cherrypick a systematic review that supports their POV. Here are two metareviews I would nominate:
  • Linde et al. Systematic reviews of complementary therapies – an annotated bibliography. Part 3: Homeopathy. BMC Complementary and Alternative Medicine 2001, 1:4; doi:10.1186/1472-6882-1-4

[3]

  • Centre for Reviews and Dissemination. "Homeopathy" Effective Health Care Bulletin 7(3) 2002 [4]
In contrast, the current ^Homeopathy & Meta analysis reference is unsystematic as it just gives the abstracts of Linde 1997 and Kleijnen 1991 (again), plus a few single trials.Jedermann 10:14, 20 September 2006 (UTC)
"Many" does not imply majority. Jefffire 11:38, 20 September 2006 (UTC)
Jefffire, "many" is wrong - only a few had OR <1 in Linde 1997. How else do you account for a combined OR >2 ? Do you have Linde 1997 in front of you as you write 'many'? In any case, labelling individual studies as +ve or -ve tends to mislead. I sense a general reluctance among some editors to acknowledge the results of a Lancet metaanalysis (your top 'authority'). Interesting that it isn't mentioned in the 'scientific' section. Thanks for the latest rv, BTW. Jedermann 12:22, 20 September 2006 (UTC)
I would strongly caution that the authors of Linde et al. 1997 subsequently published a further analysis indicating that they may have overestimated the effect of homeopathy in their initial paper (because they failed to account for rigor, I believe). It may be best to avoid that paper altogether. I would also stand more by the characterization of the vast majority of homeopathic clinical trials as 'unconvincing,' instead of 'positive' or 'negative'. T.J.C. 16:13, 20 September 2006 (UTC)

Certainly there is room for rewording to reach an amicable solution. Perhaps "a number" instead of many? On the topic of Lancet 1997, it is now supperceded by Lancet 2005 in terms of scientific rigor. Jefffire 13:05, 20 September 2006 (UTC)

Which brings us back to my original edit:
  • Although the results of most [placebo-controlled] clinical trials show a difference between homeopathic treatments and placebos,[1] many positive trials have methodological problems, and better-quality trials are more likely to give negative results.[2]
You could keep that wording, substituting 'many' for 'most' if you like (but then 'some' for 'many' later), remove Linde 1997, and it would still report current knowledge. (TJC, it is clear in the context that placebo-controlled trials are under discussion, but I've clarified it. Thanks.) As previously pointed out, Lancet 2005 found overall positive results in 110 trials, but used incremental subgroup analysis (1. better; 2. better + larger) to arrive at a group of 8 trials where the effect vanished. An NPOV article has to reflect that all the global SRs and metaanalyses (Kleijnen 1991, Linde 1997, Linde 1998, Cucherat 2001, Shang 2005) have reported overall +ve, sometimes with attenuation of effect size with better-quality trials. In fact Cucherat limited their analysis from the outset to better-quality trials (intention-to-treat only), but still got a +ve. Their method of combining p-values was weird, though. Jedermann 10:25, 21 September 2006 (UTC)
Thats roughly acceptable (with the wording present in the article at my time of writing. I get the feeling that it could be put better, but I can't think how at the moment. Jefffire 16:41, 21 September 2006 (UTC)
I've changed it to "Placebo-controlled clinical trials have given both negative and positive results". This avoids most bias of wording and simplifies it for people without medical or scientific backgrounds. Jefffire 09:13, 27 September 2006 (UTC)

Version war between Homy and Geni

  • Homy, if you want to add significant content, please discuss here first.
  • Geni, if you don't like a recent edit, stop reverting to much earlier versions that miss out subsequent corrections. Jedermann 11:28, 20 September 2006 (UTC)
What is wp:bollocks --Homy 12:27, 20 September 2006 (UTC)
WP:BOLLOCKS. Please note the 3rr rule warning I put on your talk page. It'll take a while to report you, so you have plenty of time to revert yourself. Jefffire 12:29, 20 September 2006 (UTC)

GA re-review

Nicely written article which meets all criteria except stability and for that matter, I can not grant the GA status. Other minor adjustments that can be made are :

  • It would be prefereable not to have embedded HTML links and notes in the article so, I guess, the HTML links should become notes or footnotes.
  • Make sure the C and X after the number of dilutions are capital letters. Lincher 13:07, 20 September 2006 (UTC)
It has improved a lot from what Bravada asked but there are still comments that haven't been adressed. Good luck. Lincher 13:12, 20 September 2006 (UTC)

Failure

Hmm. Lincher didn't fail it on the nomination page so it falls to me to do so, for different reasons as well as stability.

  • Most important: Length. I don't mind a good 60K article. But this is 90K ... way too long. This is almost the beginnings of the wikibooks Introduction to Homeopathy, not an encyclopedia article.

    Trim it down. Everything homeopathy-related does not need to be in this one article. I can see Criticisms of homeopathy (which might help stability), History of homeopathy and a whole series of Homeopathy by nation daughter articles easily coming out of what's here. And you might be able to come up with even more. Whatever. It does not need to be so long.

  • Intro length. Perhaps a function of the above, but even without a massive split the intro could still be trimmed to three grafs by good editing and rewriting.
  • Images. We have a picture of Hahnemann and a vijzel. The article is about homeopathy, not Hahnemann. His picture should be in there, but not in the intro.

    Homeopathy is about medicine. How about a picture of some of those preparations? A practitioner practicing? An article of this length needs pictures to break up such long blocks of text.

  • Inconsistent formatting. Those little boldface country names in "Homeopathy Around the World" should probably be separate headers, as is often done elsewhere.
  • External links. An article like this probably attracts lots of linkspamming. This section is already pretty large. Someone should look through what's there before the spam event horizon is reached.

Good luck. It appears you will have your work cut out for you. Daniel Case 22:05, 20 September 2006 (UTC)

I have tagged the article with the appropriate templates. Good to see you discussing it already. Daniel Case 22:45, 21 September 2006 (UTC)
I'm not terribly surprised. Length is certainly an issue. This is due to the bloat effect in highly-contentious articles where any removal of text putting forth ideas agreeable to one side of the argument is seen as an attack by the people who wrote it. TimVickers 22:37, 20 September 2006 (UTC)
Satellite articles would definitely help. Could we ascertain what users want? My guess is 'criticisms' and 'history' is not what they're after to begin with at least. As a patient I want to know what a therapy is, and what it will do for me and my family and friends. Jedermann 11:10, 21 September 2006 (UTC)
The main problem for me is structural. The headings don't seem logically ordered. Here are the headings from German WP: Homoeopathie
  • 1 principles 1,1 vitality 1,2 similarity principle 1.3 “provings” 1,4 dynamization 1.5 chronic disease theory
  • 2 development 2,1 history 2,2 Position in German-speaking countries 2,3 Abroad
  • 3 Trends 3.1 classical 3.2 scientific-critical (= clinical) 3,3 lay homeopathy 3,4 lay associations 3,5 veterinary
  • 4 Application 4,1 remedy choice 4,2 dosage 4,3 contraindications 4.4 aggravation/side effects 4,5 repertory
  • 5 Criticism 5.1 no proof of efficacy 5.2 no plausible mechanism 5.3 internal contradictions 5.4 dangers 5.5 miscellaneous criticism
  • 6 footnotes
  • 7 related therapies
  • 8 literature 8.1 sources 8,2 secondary literature 8.2.1 supporting 8.2.2 critical 8.3 scientific metaanalysis of effectiveness

It works better for me. Do we have anything like German section 4? I also prefer their clear bibliography and references. Our article is overloaded with refs at present, many of which are dubious.

The French site is also interesting. The Dutch have the best evaluation section, as they list all the global analyses. Jedermann 13:30, 21 September 2006 (UTC)

I've not objection to reordering, but we do need to bear in mind that the scientific dispute over homeopathy is rather a notable part of its story. I've also no obection to subarticles on the history of the subject, but a subarticle on criticisms would become a PoV fork (eithere pro or con) and would cause more trouble in the long run. I recommend against that very strongly. Jefffire 16:41, 21 September 2006 (UTC)

Whatever WP:POVFORK says, I submit that this article is what inevitably happens when you avoid POV forks for the sake of avoiding POV forks. I feel that as long as a "Criticism(s) of ..." presents any responses to the various criticisms and sources for those responses and criticisms (and is adequately policed for any violations of those guidelines with in-text comments and one or two vigorous watchers), it is clearly the lesser of two evils. Daniel Case 22:45, 21 September 2006 (UTC)
I still have this article on my watchlist as I am curious how this discussion is going to develop, but I feel I have to butt in and support Jefffire on the fact that it would amount to a POV fork. Funnily enough, while I am not a fan of long articles on GA, the length did not strike me while I was reviewing it, I had no idea it was a whooping 90K! It all seems quite focused on the topic and interesting, so I believe it is better to keep working on it in its present form, and perhaps think about splitting and such once a fairly stable version is reached. Bravada, talk - 17:01, 21 September 2006 (UTC)

Lancet 2005

If I can interject here, as I wrote the original wording of the Lancet 2005 study, and as some issues here seem to have come up again and again in the past. The Lancet study is not a conventional meta-analysis; but a novel approach that considered the hypothesis that publication bias leads to an excess reporting of positive results from small trials. The authors analysed 110 trials and 110 matched conventional trials, the drug effects diminished with trial size and rigor in the homeopathy group but not in the conventional medicine group, leading them to conclude that the results are consistent with all the effects of homeopathy being placebo/false positive results. The statement that this is the largest and most rigorous I think came initially from the accompanying Lancet editorial, I'm not sure now, lost in a history of edits and revisions that has introduced distortions.

The study has been widely misunderstood, not surprisingly because it has an unusual approach; one misunderstanding is perpetuated in the present wording (8 trials; not true).

I think my original wording of this section was: In August 2005, The Lancet published a meta-analysis of trials of homeopathy, the largest and most rigorous analysis so far, involving 110 placebo-controlled homoeopathy trials and 110 matched conventional-medicine trials [3]. The outcome of this suggested that the clinical effects of homeopathy are likely to be placebo effects. For reactions to this study from homeopaths, see [5] [4] [6]’’

I still think that’s all that need or should be said. However in response at different times and different people, the section was expanded to clarify the study (apparently using my account from the Talk pages, but this was edited and errors introduced later. I think it’s best to see words from the article;in Table 3 of the paper for instance, the legend is: Univariable meta-regression analysis of treatment effects in 110 placebo-controlled trials of homoeopathy and 110 matched trials of conventional medicine. (The paper also compares the very best studies of which there are relatively few, and that's where the 8 comes from). In the discussion "We compared the effects of homoeopathy and conventional medicine that are seen in placebo-controlled trials, examined the presence of bias resulting from inadequate methods and selective publication, and estimated results in trials least affected by these biases. We assumed that the effects observed in placebo-controlled trials of homoeopathy could be explained by a combination of methodological deficiencies and biased reporting. Conversely, we postulated that the same biases could not explain the effects observed in comparable placebo-controlled trials of conventional medicine. Our results confirm these hypotheses: when analyses were restricted to large trials of higher quality there was no convincing evidence that homoeopathy was superior to placebo, whereas for conventional medicine an important effect remained. Our results thus provide support for the hypothesis that the clinical effects of homoeopathy, but not those of conventional medicine, are unspecific placebo or context effects."

What I suggest is that any account of the Lancet paper should be firmly based in what the authors actually said, anything else risks OR. We report, we don't interpret

On the Linde study, this is complex as the authors themselves complained subsequently about their findings being misrepresented (by both sides in fact) and they appeared to back off. It might be worth now backtracking citations to that study to see how it has lasted, but I think it has indeed been superceded Gleng 15:33, 20 September 2006 (UTC)

Well what a useless waste of people's time the whole thing was. Forsure, I for one will not be visiting this article again. Too much bitching over hairsplitting details and too much bile between so-called editors. Count me out. Peter morrell 17:38, 21 September 2006 (UTC)
  • Comment Agree. The Lancet article is widely criticized. Homeopathy has developed its own empiric "science" since 200 years and the current "homeopathic clinical trials" have not much to do with homeopathy. --Homy 07:40, 22 September 2006 (UTC)

Who cares what you think? You've had an opportunity here to learn how homeopathy's reported positive effects over placebo are nonsense and are almost certainly due to publication bias. But your mind is closed to everything except your "beliefs". Go back to your cauldron and crystals.

Where do we go from here?

I propose separating several spin-off articles. I think "Clinical testing of homeopathic treatment" and "History of homeopathy" are good suggestions from earlier. What do people think? TimVickers 20:09, 21 September 2006 (UTC)

As I say above, I fully support "History of homeopathy" as a sub-article as it seems a fascinating subject that there is a lot to write about. I have to say that at this time I would recommend any "Clinical testing of homeopathic treatment" for deletion if it were created as based on similar article I have seen, such as "Objective validity of astrology" or "Cold fusion controversy", it would just degenerate into edit wars and/or a PoV fork. On the other hand, "Creation-evolution controversy" seems to work, but it does have a veritable army of dedicated and experienced editors to keep it in line. Jefffire 20:35, 21 September 2006 (UTC)

So you think the non-controversial parts of this article are the best ones to put into separate articles? TimVickers 21:10, 21 September 2006 (UTC)

This is a very good rule of a thumb for such splits. Bravada, talk - 22:34, 21 September 2006 (UTC)

As I have said before: Contents points 2 and 3 belong in classical homeopathy esp. "Miasms as a cause of disease" is no general subject, but with specific knowledge --Homy 07:12, 22 September 2006 (UTC)

Having suggested 'History', I now think it's too big on its own. What about a number of basic topics to include their own histories? Like 'Homeopathic pharmacy' (manufacture, dosages, forms, table of potency dilutions, Hahnemann's repeated changes in methods, standard pharmacopeias, and some rather remarkable methods of manufacture listed here)? And the article Homeopathic proving is already there but could use material from here about origins and predecessors, how Hahnemann did it, and what has happened since. The head article surely only needs a brief description, and some reference to systematic reviews of recent provings (highly critical). I understand Jeff's caveat about a science section, but it's such a large complex topic. Can I suggest that here it should use only reviews (such as 2 that I posted done by neutral reviewers). The same can be done for basic research - I know at least 3 systematic reviews. Even if you don't accept this, Scientific testing needs a rewrite. It's off-balance with disconnected subheadings and items that belong under higher-level headings given their own section (2006 cancer review, basophil) then a section that tries to explain what EBM is before listing unsytematically chosen reviews. I also don't understand the need for the Vaccination section, some of which belongs under Antivaccinationist. Anyway, there is too much fascinating information waiting to go into satellite articles that won't fit in here. Jedermann 11:35, 26 September 2006 (UTC)

one sided article

Homeopathy is a complete and utter scam. how about a section on all of the people who are now dead because they chose to use homeopathy rather than actual medicine to treat their illnesses? how about a section on the many many studies that have debunked this snake oil treatment? How about the many suicide attempts made by skeptic organisations around the world where individuals have swallowed entire bottles of homeopathic sleeping pills and other snake oil remedies? How about statement of the fact that some of the dilution levels claimed by homeopathy are completely impossible to attain? how about explaining the amount of medical training required to create homeopathic remedies or to dispense them?

This article is completly biased towards the multi-billion dollar snake oil industry. 194.46.231.96 22:07, 23 September 2006 (UTC)

Thanks for that, 194.46.231.96. Your user Talk adds great weight to your argument: http://en.wikipedia.org/wiki/User_talk:194.46.231.96 Jedermann 14:22, 24 September 2006 (UTC)
Just as your choice to argue via the Ad hominem fallacy adds weight to yours. Is that your best argument? Skeptic Jim 08:29, 26 September 2006 (UTC)
Anyone who bothered to read the article beforehand would not aggressively demand things that are already included. And pointing to User:194.46.231.96's astonishing record of WP vandalism and abuse was a warning not an argument, since this is a contentious article that has been repeatedly vandalized. As for me, I'm not in an argument for or against. I would just like proponents and skeptics to respect such facts as can be ascertained in the present state of knowledge, and to improve the quality of the criticism they aim at each other. Jedermann 09:18, 26 September 2006 (UTC)
Well that was an ad hominem attack regardless. I didn't read the full article because it's way too long. But I did read the first three quarters of the introduction and I didn't see anything that in any way suggested that homeopathy is considered to be a pseudoscience, bullshit, snake oil or even unscientific by the scientific community. When you have a 90k article I think it's unreasonable to assume people will read the entire article Skeptic Jim 21:58, 26 September 2006 (UTC)
If you had read ONLY the introduction it would have been clear to you it was a balanced article. I can't understand why you would read the first three quarters of the intro and then stop right before reading the paragraph that starts with the word Critics. Why did you choose not to read that paragraph? This article is about homeopathy as such it focuses on the practice and history of homeopathy. Why would you expect it to be foremost an exposé of homeopathy? David D. (Talk) 03:07, 27 September 2006 (UTC)

"Critics of homeopathic medicine frequently describe it as pseudoscience[12] and quackery.[13] The theory of homeopathy is inconsistent with known laws of chemistry and physics, since it states that extreme dilution makes drugs more powerful by enhancing their "spirit-like medicinal powers."[14] Although many placebo-controlled clinical trials show a difference between homeopathic treatments and placebos, most trials have methodological problems, and better-quality trials are more likely to give negative results.[15] Additionally, cases have been reported of life-threatening illness resulting from attempts to treat serious conditions solely with homeopathic remedies.[16][17]"

Is this paragraph good enough? I've tried to get the scientific viewpoint across in a way that is acceptable to the other editors. - CronoDAS 01:42, 27 September 2006 (UTC)

Good points all round. I think I will steer clear of this article in future since the topic tends to make my blood boil and the last thing we need is emotive editors. If I happen to stumble on sources I think relevant I will post them here and let a less emotive editor review its content. Skeptic Jim 10:03, 27 September 2006 (UTC)


Topics box query

How do you edit the Topics box at the bottom of the article, where there are a number of errors?

e.g. Anthroposophy should be Anthroposophical Medicine, Toxicology isn't a therapy (probably means Homotoxicology) Jedermann 14:14, 24 September 2006 (UTC)

Numbers of people receiving homeopathic treatment

In the category "Numbers of people receiving homeopathic treatment", it says that 56% of population in belgium uses homeopathy. This information is based on this study. Anyway, in the same study, it says that only 31% of people in Belgium uses "Any form of complementary medicine", where it includes homeopathy.

I'm just wondering. Is it that in Belgium homeopathy is not "a form of complementary medicine" and that's why the percentage of people in homeopathy is bigger? Or what is the reason for this?

I don't know, but here's a possibility: the term "complementary medicine" is sometimes used to mean "alternative" or "traditional" treatments used alongside scientific medicine, rather than instead. So if 25% of Belgians use homeopathy exclusively (i.e. do not use scientific medicine at all) and 31% use homeopathy as well as scientific medicine, then the 56% figure could come from there. --FOo 20:18, 30 September 2006 (UTC)
Thank you, I hadn't thought in that! Anyway, 25% of Belgians not using any kind of scientific medicine at all... sounds pretty much like a lot, right? It would also mean that at least 10% of british, 23% of swedish and 4.8% of danish would also not use any scientific medicine. How can this be verified? or can anyone else give any other explanation for those numbers?
The report states that "homoeopathy will be considered alternative whoever prescribes it". From the context, I take this to mean regardless of which practitioner and nationality is under consideration. The report also starts out by using the term "alternative medicine" and then goes on with discussing "complementary medicine" after referring to a book by the British Medical Association. It appears to me that they use the terms more or less interchangeably. I must admit that I also find it difficult to interpret the numbers given in the table posted in the article. My first impulse was to read the first colum as the percentage who used _any_ of the following complementary/alternative treatments of column 2, 3, 4 etc. Thus, 69% (100%-31%) of belgians would then not use any complementary medicine. Of the 31%, 56% would use homeopathy. The problem with this interpretation is that column 2-5 add up to more than 100%. This might in turn be explained by some users of complementary medicine using multiple remedies, i.e. checking off both homeopathy and osteopathy in the questionnaire. But since this is not adequately explained in the report, I tend to think that it should be replaced by a study that defines the values it reports in a better way.
As for the percentages of the population of Belgium, England and Sweden not using any scientific medicine at all, I agree that values you mention above sounds very, very high. EthicsGradient 14:01, 24 October 2006 (UTC)