http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/ in health, homeopathy, and life en john@vitalpatterns.net Copyright 2008 by the respective commenters 2008-04-04T06:58:00-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#91 John - just a quick thought. You might like to look at the modalities for researching psychotherapy or psychiatry. The long term patient/therapist relationship would appear to have much in common with the patient/homeopath relationship you describe above. Unfortunately, as far as I’m aware, the reliability of evidence is not very good - with consequent dispute over the efficacy of psychiatric treatments or existence of conditions and division between various schools of practitioners. Such research wouldn’t silence the sceptics (but you wouldn’t care about that anyway) but it might be very interesting. {categories backspace="1"}{category_name}, {/categories} 2007-12-09T11:00:21-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#96 But homeopathy is not therapy. The patient-homeopath relationship is a factor that influences the homeopath’s ability to gather the data needed to find a similar remedy, but that relationship does not itself cause the result. And we don’t try to build a long term relationship in the way a therapist might: the ideal is to see the patient a few times and be done. {categories backspace="1"}{category_name}, {/categories} 2007-12-09T20:06:34-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#101 John - sorry, I obviously didn’t make myself clear. You seemed to be suggesting that homeopathy couldn’t conform to ‘reductionist’ tests (condition X is treated by remedy Y) because a homeopath will provide different remedies to each patient. My point was that such a context is already handled by existing research methods. The reference to psychotherapy was by way of analogy. There, each threapist/patient relationship is unique and individual treatments will be tailored to the specific needs of the patient. If ‘reductionist’ double blind PCTs are inappropriate for homeopathy then you might want to look at other research modalities which are designed to assess a similar context to that you describe. {categories backspace="1"}{category_name}, {/categories} 2007-12-09T22:21:38-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#104 Conventional clinical trials test the action of a certain medicine, but in homeopathy, it’s really the similarity that is being tested, not the substance itself. Clinically, the similarity is the hypothesis, if you will, and the results the evidence. The similarity as the hypothesis is an interesting view. I think that should be the line of thinking to develop a test that can prove the working of homeopathy. And the results as the evidence is even a greater challenge, as it would have to take in account the direction of cure. {categories backspace="1"}{category_name}, {/categories} 2007-12-11T01:19:46-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#105 I seem to have confused things with my statement, “Clincially, similarity is the hypothesis...” The word “clinically” is very important there. With an individual patient, the similarity is effectively what we are testing, because we assume (know from experience) that a similar remedy will work and a more similar remedy will work better. I already know how to judge this, although with an individual patient there are levels of uncertainty. But from the point of view of a conventional trial, where you are trying to test “does it (homeopathy) work at all?” then the similarity issue is something you have to manage in your experiment. The straightforward way is to have homeopaths individualizing remedies as part of your trial, instead of giving remedy X to everybody. But then you need several homeopaths and numerous remedies involved, which is doable but more complicated. {categories backspace="1"}{category_name}, {/categories} 2007-12-11T05:59:15-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#106 Hello, I’ve been reading & commenting on a few of the ‘homeopathy blogs’ with interest. However, on one blog, freetochoosehealth, almost all my comments are being deleted, however polite. Today I have had two posts deleted that were just saying ‘Happy Christmas’. This confused me. I think that everybody should be free to make choices about their health based on all the best information we have. Censoring comments suggests an unwillingness to engage in debate. If you hold strong beliefs, you should be prepared to defend them with argument and evidence, or re-evaluate them if you can’t do so. Censorship doesn’t help anyone. I’ve put up a blog of my own, just containing the titles & a link to each of FTCH’s posts. I won’t delete comments on there (if I have no choice, I will always explain why), and anyone from any side of the debate is welcome to write whatever they want, without censorship. I would rather hope that you can all manage polite & reasoned discussions, but I won’t delete what’s not. Please join me, and submit comments on both freetochoosehealth and http://freefromcensorship.wordpress.com/ Many thanks, ‘Duck’. {categories backspace="1"}{category_name}, {/categories} 2007-12-11T12:10:11-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#107 I don’t appreciate the blog spam, and I really don’t want to get in the middle of a fight between two other bloggers, but I will leave Duck’s comment up. I have read and heard about a lot of nasty and even slanderous things that have been written online and even in British newspapers. Basically mean attacks that are put up in the guise of “debate.” For some so-called skeptics this is not a debate, but a kind of game. I don’t know if this is the situation with Duck and FreeToChooseHealth or not, and would prefer not to judge it. But I do want to say clearly that I will remove nasty or slanderous comments from my blog, and may decide to ban someone who has made them. I would also like to take this opportunity to thank the people who have commented on this blog. It has in fact been a good, civilized debate. Thank you. [Well, at least it was at the time I wrote this. But still, thanks to those who have actually debated.] {categories backspace="1"}{category_name}, {/categories} 2007-12-12T09:14:00-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#129 With an individual patient, the similarity is effectively what we are testing, because we assume (know from experience) that a similar remedy will work and a more similar remedy will work better. If you don’t know what the “similarity” is in any one case, which you don’t because, as you say, you are testing it in every case, how do you know (from experience) that “a similar remedy will work and a more similar remedy will work better”? I know that the law of similars is one of the guiding principles of homoeopathy, but if there are no hard and fast definitions of similarity (because of the added variability of the particular patient in that particular situation) then you run the risk of defining “the most similar” simply as the remedy that appears to have been most effective. This leads you into the problem of a circular definition. More problematically, it can lead to the treatment method simply being the prescription of various preparations in no hard and fast order, and then the retrospective attribution of efficacy to whatever preparation the patient was taking when their condition resolved. {categories backspace="1"}{category_name}, {/categories} 2007-12-17T07:36:05-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#132 This leads you into the problem of a circular definition. Here is the situation for any single patient. Assumption: The law of similars 1. I give what I think is a similar remedy. 2. It works. 3. I conclude that it was in fact similar. If I used this as proof of the law of similars would in fact be circular. But the general case used to validate the homeopathic method is different. 1. I use a method based on the law of similars. 2. It works, i.e. I get significantly better results than with my current method or a placebo method. 3. I conclude that the Law of Similars is valid. Now, assuming the law of similars is valid, I can use it clinically to evaluate similarity. The issue is the “it works” part. All I’m saying here is that if you are going to test whether it works or not, you have to consider the similarity issue when you are designing your test. Otherwise you end up with this: 1. I use some method that is kind of like but isn’t homeopathy. 2. It doesn’t work. 3. I conclude homeopathy doesn’t work. That is a logical falacy. {categories backspace="1"}{category_name}, {/categories} 2007-12-17T19:50:08-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#137 But the general case used to validate the homeopathic method is different. 1. I use a method based on the law of similars. 2. It works, i.e. I get significantly better results than with my current method or a placebo method. 3. I conclude that the Law of Similars is valid. I can’t see a problem with this, but it doesn’t fit in with your earlier statement that Clincially, similarity is the hypothesis...” With an individual patient, the similarity is effectively what we are testing If it is genuinely a “law” of similars then there is no test to be done, because you give the remedy that causes similar symptoms and the patient recovers. It strikes me that what you’re giving yourself freedom to do is observe that the first-line remedy (that primarily suggested by the “law") doesn’t work and then restrospectively saying “well that remedy wasn’t really similar, because this patient is also old/stressed/young/pregnant/etc.” This is hardly logical or rigorous (it may be an example of the “No True Scotsman” fallacy). Moreover, there seem to be few comparisons of this method against genuine placebo (and defining ineffective remedies as “placebo” isn’t valid). {categories backspace="1"}{category_name}, {/categories} 2007-12-18T02:14:10-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#138 “It works, i.e. I get significantly better results than with my current method or a placebo method” How do you know the results are better than (a) your current method and (b) placebo? {categories backspace="1"}{category_name}, {/categories} 2007-12-18T02:15:55-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#148 “Don’t get me wrong: I’m not saying that we can’t test any claims about homeopathy.” Good, that’s excellent. Could you elaborate about what specific claims can be tested, and what sort of test would be appropriate in each case? {categories backspace="1"}{category_name}, {/categories} 2007-12-19T06:21:42-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#165 If it is genuinely a “law” of similars then there is no test to be done, because you give the remedy that causes similar symptoms and the patient recovers. But the problem is that it is very difficult to see the similarity, because you only see a part of the total picture. The similarity can be seen as two puzzles, the real one and the mirror. Sometimes you have many pieces from the mirror (remedy) and just a few from the real one (patient). Or the other way around. Or just a few from both. So the only way to recognise them as similar is to have a mental picture of the complete puzzle. {categories backspace="1"}{category_name}, {/categories} 2007-12-20T22:54:11-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#175 The similarity can be seen as two puzzles, the real one and the mirror. Have you seen Scholten’s presentation on the different ways that homeopaths match remedies, using a painting with only parts of it exposed? It is a great metaphor. I hope to put something like it up here eventually. {categories backspace="1"}{category_name}, {/categories} 2007-12-22T09:01:09-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#176 As for the tests, there seems to only be one test that matters to skeptics: is there more to homeopathy beyond placebo effect? We went round on that one in this blog on my How Do You know it works? post. You can go read up on the existing evidence on the NCH’s website or Dana Ullman’s website. And if that’s not enough (obviously it isn’t to skeptics) then I cannot produce any more at this time. In any case, the point of this post was to say that if you don’t account and control for the similarity effect, your experiments will be useless because you haven’t actually tested homeopathy, only a claim about potentized substances that isn’t the same as the one that homeopathy is making. Along those lines, it’s also important to know that not every product that claims to be homeopathic (nor every practitioner) is actually so. See my definition of homeopathy here. Just because you use a potentized substance doesn’t make it homeopathy. {categories backspace="1"}{category_name}, {/categories} 2007-12-22T09:14:31-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#180 Have you seen Scholten’s presentation on the different ways that homeopaths match remedies, using a painting with only parts of it exposed? I have not seen this presentation, but I think it would be very interesting. Is it something he used at a seminar? {categories backspace="1"}{category_name}, {/categories} 2007-12-22T23:00:04-08:00 http://woowooscience.com/index.php/blog/post/testing_homeopathy_and_the_similarity_problem/#204 I have not seen this presentation, but I think it would be very interesting. Is it something he used at a seminar? Yes, it was a graphical attempt to explain/justify family or group analysis for selecting remedies, and how it compares to the more standard method. {categories backspace="1"}{category_name}, {/categories} 2008-01-10T10:30:27-08:00