Boiron, the world’s largest manufacturer of homeopathic products, has recently been in the headlines more often and less favourably than their PR-team may have hoped for. Now they have added to this attention by publishing a large and seemingly impressive multi-national study of homeopathy.
Its objective was “to evaluate the effectiveness of homeopathic medicine for the prevention and treatment of migraine in children”. For this purpose, the researchers recruited 59 homeopaths from 12 countries who included into the study a total of 168 children with “definite or probable” migraine. The homeopaths had complete freedom to individualise their treatments according to the distinct characteristics of their patients.
The primary study-endpoints were the frequency, severity and duration of migraine attacks during 3 months of homeopathic treatment compared to the 3 months prior to that period. The secondary outcome measure was the amount of days off school. The results are fairly clear-cut and demonstrated that all of these variables improved in the period of homeopathic care.
This study is remarkable but possibly not in the way Boiron intended it to be. The first thing to notice is that each homeopath in this study barely treated 3 patients. I wonder why anyone would go to the trouble of setting up a multi-national trial with dozens of homeopaths from around the globe when, in the end, the total sample size is not higher than that achievable in one single well-organised, one-centre study. A multitude of countries, cultures and homeopaths is only an asset for a study, if justified through the recruitment of a large patient sample; otherwise, it is just an unwelcome source for confounding and bias.
But the main concern I have with this study lies elsewhere. Its stated objective was “…to evaluate the effectiveness of homeopathic medicines…” This aim cannot possibly be tackled with a study of this nature. As it stands, this study merely investigated what happens in 3 months while children receive 3 months of homeopathic care. The observed findings are not necessarily due to the homeopathic medicines; they might be due to the passage of time, the tender loving care received by their homeopaths, the expectation of the homeopaths and/or the parents, a regression towards the mean, the natural history of the (in some cases only “probable”) migraine, any concomitant treatments administered during the 3 months, a change in life-style, a placebo-effect, a Hawthorne-effect, or the many other factors that I have not thought of.
To put the result of the Boiron-researchers into the right context, we should perhaps remember that even the most outspoken promoters of homeopathy on the planet concluded from an evaluation of the evidence that homeopathy is ineffective as a treatment of migraine. Therefore it seems surprising to publish the opposite result on the basis of such flimsy evidence made to look impressive by its multi-national nature.
I have been accused of going out of my way to comment on bogus evidence in the realm of homeopathy. If this claim were true, I would not be able to do much else. Debunking flawed homeopathy studies is not what I aim for or spend my time on. Yet this study, I thought, does deserve a brief comment.
Why? Because it has exemplary flaws, because it reflects on homeopathy as a whole as well as on the journal it was published in (the top publication in this field), because it is Boiron-authored, because it produced an obviously misleading result, because it could lead many migraine-sufferers up the garden path and – let’s be honest – because Dan Ullman will start foaming from the mouth again, thus proving to the world that homeopathy is ineffective against acute anger and anguish.
Joking apart, the Boiron-authors conclude that “the results of this study demonstrate the interest of homeopathic medicines for this prevention and treatment of migraine attacks in children”. This is an utterly bizarre statement, as it does not follow from the study’s data at all.
But what can possibly be concluded from this article that is relevant to anyone? I did think hard about this question, and here is my considered answer: nothing (other than perhaps the suspicion that homeopathy-research is in a dire state).
15 Responses to Boiron’s new study of homeopathy
Leave a Reply to m Cancel reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
The point is surely that, apart from the placebo effect itself (which can be done with plain water and coloured sugar pills), there is northing to research. Finding positive results in homeopathy research is the art of wishful thinking and writing fiction.
More interesting research would be to investigate why people continue to assert black is white after black has been shown incontrovertibly to be black, and white has been demonstrated beyond reasonable doubt to be white.
I think the answer lies in companies like Boiron, Nelso, Boots etc., having a vested financial interest in taking advantage of a public desire to defer to authority. For example, Boots sell homeopathic preparations because there is a demand. There is a demand because Boots sell them. Boots are a trusted authority on all things medical, apparently, and the public believes Boots wouldn’t sell something they knew to be ineffective or even fraudulent. Boots have an interest in maintaining the deception. Boiron have even more of an interest.
It’s almost impossible to escape the conclusion that all the big players in homeopathy are perfectly well aware it is a massive deception. But homeopathy is legal so it’s ok to decieve and take advantage of public ignorance.
I agree, but this vicious circle is also perpetuated by the Ullmans of this world – who, of course, have a vested interest as well but seem to be firm, quasi-religious believers, thousands of patients who are unable to differentiate between specific and non-specific therapeutic effects, and seemingly altruistic, voiciferous half-wits like prince charles.
Just the latest in a long list of trials of homeopathy by Boiron. They have been heavily criticised for the poor standard of their trials. It would be less costly for them to produce fewer trials of high quality. Their motives for behaving in this fashion have to be seriously questioned.
If even a review of Mr Edzard in 2004 when evaluated in homotoxicology remedies. Ernst acknowledges that the studies were good quality, but unexpectedly finds no credible studies and states that “no evidence”. That is the outcome of this “systematic review” not derived from the premises.
I assume that you are referring to this. What part of “important flaws were found in all trials. These render the results of the primary studies less reliable than their high Jadad scores might suggest” do you not understand?
Yes, im refferring this study.
Interestingly Mr. Edzard always says the same. All systematic reviews I’ve read say they “found serious methodological flaws.” All reviews are entirely narrative and have a high bias of subjectivism. The party manages to fire Edzard studies have efficacy over placebo is the argument of the “implausibility”:
1. According to Edzard, homeopathy contradicts all laws of physics. This argument is similar to that used by David Grimmes in a recent article (low quality, who the hell agreed to publish it?) In FACT. Both Grimmes Edzard and do not show all these laws that say contradicts homeopathy. I think this myth was partly physical product of Robert L. Park. It is within this category that rejects Edzard studies.
2. The “major methodological flaws” are fun because Ernst tests are mostly of poor quality. Do not even have reliable statistics or anything. Not compare even with the flawed analysis of Shang and cherry picking. et.al.
3. Some, not all, Ernst trials have only as the author. What power bias in the analysis!
4. Additionally I want to mention that one of his criticisms of the Swiss report (in FACT) is that it was done by homeopaths and therefore is more tending to bias. The most fun of his critique is that only focuses on the clinical trials, ignoring the part of physical and biological studies. In order to give the appearance that you have evaluated all the Swiss report. And most bizarre is that he dares to say that homeopathy is not credible “implausible”. To this end cites an article in Biologist that repeat.
Finalmente, por que el meta analisis de Shang no incluyo los estudios clínicos de Ernst?
Any response?
M: will you pay for the Sanish course I need to take for being able to understand your question?
I assumed from your description that the effect could be entirely explained by regression to the mean, but I failed to account for how badly designed the study is. At this level of bad design one has to assume that it is designed by the marketing department and not by anyone interested in curing migraines. The journal should be ashamed.