The UK ‘Society of Homeopaths’ (SoH) is the largest professional organisation of UK non-doctor, so-called lay- homeopaths. On their website, the SoH made very specific claims about homeopathy; in particular, they listed conditions for which homeopathy had allegedly been proven to be effective. These claims have now thoroughly been debunked, and the evidence the SoH produced in support of their claims has been shown to be misleading, cherry-picked or misinterpreted.
I have no idea who conducted the above-named investigation and made a youtube video of it, but I think it is essentially correct and well worth watching. My own experiences with the SoH relate mainly to two encounters.
The first was a complaint I made about one of their high-ranking officers, Ralf Jeutter. He had been promotiong homeopathic vaccinations on his website (needless to stress, I think, that there is no evidence to support the notion that homeopathic vaccinations are effective). As I felt that the SoH dragged their feet pursuing my complaint, I had to send several reminders. Eventually, they considered it and concluded that Reuter had done nothing wrong. This, presumably, is the reason why, even today, he can state on his website that Homeopathy is used to help individuals in dealing better with kinds of infections such as leptospirosis, meningitis and cholera. All is fine, it seems as long as a disclaimer is added: Any information obtained here is not to be construed as medical OR legal advice. The decision to vaccinate and how you implement that decision is yours and yours alone. The evidence for the efficacy of homeopathic immunisation is ‘anecdotal’. That means it is based on individuals’ reports past and present.
My second encounter with the SoH relates to my 2010 analysis of the SoH code of ethics and their adherence to it. The code demanded that:
- ‘all speculative theories will be stated as such and clearly distinguished’
- ‘no advertising may be used which expressly or implicitly claims to cure named diseases’
- ‘Advertising shall not be false, fraudulent, misleading, deceptive, extravagant or sensational.’
Encouraged by these assurances, I decided to study the websites of some members of the SoH, and soon discovered numerous and very obvious violations of the above-mentioned imperatives. In an attempt to find the root of these transgressions, I scrutinised the SoH’s own website where I found a multitude violations on all levels of the SoH’s own code of ethics. Many of the violations related to claims which were not supported by evidence. In other words, the largest professional UK organisation of lay- homeopaths misled the public in several rather devious ways:
-
they pretended to adhere to a code of ethics which forbids members to mislead the public
-
SoH -members nevertheless did mislead the public in ways that public health at risk
-
and they did so not least because the SoH followed exactly the same strategy
-
thus the SoH violated its own code of ethics to the detriment of public health.
My analysis was conducted a while ago, and some might hope that the SoH has stopped systematically misleading the public. This hope, however, is harshly disappointed when you watch the brand-new video entitled TESTING HOMEOPATHY mentioned above. As the SoH is about to celebrate 35 years of wisdom, courage, knowledge and prosperity, I do wonder whether this should not be 35 years of dangerously misleading the public.
What do you think?
30 Responses to Drowning in a sea of misinformation. Part 2: The UK ‘Society of Homeopaths’
Leave a Reply to George Cancel reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
Only today I received a response to a complaint I made a few months ago. I have had several responses, none of which adequately addressed my complaint. These two paragraphs from my reply should give an idea of what I foind unacceptable:
‘Your reply does not address a substantive part of my complaint, which is that you do not require your members to comply now with your own existing Code of Ethics, which (paragraph 39) states: “No promise of cure, either implicit or explicit,should be made of any named disease.”
‘This is completely independent of your deliberations with CAP/ASA and I am left with the inescapable conclusion that you and your predecessor have merely been engaging in diversionary tactics and that the Society of Homeopaths has no intention of taking my complaint seriously. What is the point of your Code of Ethics if your members are not required to comply with it? This document is supposedly intended to give the public confidence that you require your members to behave ethically. In my opinion, the public deserves to be informed that your members are free to pick and choose which bits, if any, to comply with.’
’35 years of wisdom, courage, knowledge and prosperity’
The video by CHL is a very good presentation showing that homeopaths do not understand the concept of evidence.
The above ludicrous statement demonstrates an inability to understand the English language.
I think it’s a travesty.
It’s the same here–market anything you like as long as you put a disclaimer to the effect that the FDA (Food and Drug Administration) has not approved the product. Such a “curse” matters little the woo-afflicted because part of their conspiracy mongering is that the FDA is even more evil that BigPharma and is not to be trusted to begin with.
All is well in the eyes of the FDA unless deaths start to show up. Unfortunately, the deaths that occur from avoidance of “mainstream” medical care don’t get counted by any agency.
Ah, the FDA Paradox: The evil FDA is controlled and funded evil by Big Pharma and authorise their evil products without any proper testing, yet the FDA also gives homeopathy and supplements a free ride…
Dr. Ernst
I did not understand what they were talking about in the video..
How, for instance , this http://www.ncbi.nlm.nih.gov/pubmed/12634583 is debunked and by whom ? From a commitee who decided that even if the results were favorable to homeopathy, the study is invalid only because is about individualized homeopathy ? Disregarding all the evidence and the fact that this a peer review paper published in a mainstream journal ?
The study concluded
“The results from these studies confirm that individualized homeopathic treatment decreases the duration of acute childhood diarrhea and suggest that larger sample sizes be used in future homeopathic research to ensure adequate statistical power. Homeopathy should be considered for use as an adjunct to oral rehydration for this illness.”
Is this video evidence of scientific or logical thinking ? The most scary of all is the 300 likes – I know rational sceptisism is cool and/popular but unfortunately incompatible with logic…
If you think that peer-reviewed papers published in mainstream journals are inevitably correct, what do you do when, as sometimes happens, they contradict each other?
The fact that this childhood diarrhea research was financed by a grant from Boiron is enough to call for a high degree of suspicion
Do you mean that all the research funded by big pharma on vaccines or on other medications are suspicious as well?
By the way, George, did you understand the comments on the childhood diarrhoea papers here?
If papers contradict each other I m becoming curious to learn more and not dismissive based on my superstitions and beliefs..
I have replied to the comment you said – it is really naive…
In another thread on this blog I did explain why the Jacobs-studies on childhood diarrhea are bogus.
Note: The author of the study you reference (J Jacobs) just summed up work of her own. So this is not an independent review.
Sorry George: Jacobs’ work on childhood diarrhea is bogus.
Those children suffered from diarrhea and dehydration. Two therapies were applied starting at the same time: homeopathy and WHO-rehydration. The big step occurred in both groups at day 1 in similar magnitude, so could not be attributed to homeopthy. After that, verum group actually recovers at the same speed, if not just a little more slowly than placebogroup, but had a slightly better start, which well could be a consequence of the mismatch of the groups in age and size. The only point in time when the difference in groups reached significance was at day 3, not before and not after.
See fig 1 and table 2 in the study conducted in Nicaragua.
http://amcofh.org/sites/default/files/Acute%20Care%20Class%201%20Research%20JJacobs.pdf
Unfortunately Jacobs decided not to give the details of the patients’ progress in her further trials nor in her meta-analysis, but what can be seen frm the data, they look pretty similar.
I don’t know what is bogus – the study or you speculations?
1. WHO-rehydration does NOT reduce the total duration of the symptoms – that was they were trying to measure –accurately.
2. They have sophisticated ways of measuring these effects – saying a little bit or slightly is OK for the layman but researchers base their results on their EXACT measurements using statistical methods usually unavailable to people who have not study the subject so your objection : “just a little more slowly than placebogroup, but had a slightly better start, which well could be a consequence of the mismatch of the groups in age and size.” is kind of naive……..By this mode of thinking -for every trial with positive or negative outcome in any filed one could have been pointed out different undetected factors ( individual response to medication, weight, alcohol , smoking let alone ….fairies….) The meaning of a control trial is to eliminate the possibilities with precision and not to say –a little bit , not too much…
Regarding the later study which agains showed homeopathy s efficasy , the result I significant and the USA objection that it might be due to other factors is really naive -: for every study a critic could claim that some unmeasured factor would be the real cause…
It really fallacious to regard every study on homeopathy with positive results as wrong because evidence for homeopathy really conflicts with your beliefs about the method – Everybody in research knows that this is the worse mode of thinking for a scientist ..Why for instance don’t you object the design of non individualized homeopathy trial which found that it was placebo?? These could not be due to other factors ?
Mojo regarding the critisn on the earlier study I don’t know what AMA meant – My guess is that individualized homeopathy was their concern since it gives a different remedy to patients ..But I’m not sure …
I posted an answer to your comment in the thread about the worst platitude of all, but I see you posted the same comment here. So, for the sake of clarity, lets stay on this thread here.
So let me give you some data first:
In the Nicaragua trial – for the others such data are not available – the fig 1 gives the following numbers of stools at day 1, day 2 etc.:
Verum: 7.9 / 3.4 / 2.8 / 2.1 / 2.0 / 1.6
Placebo: 7.5 / 4.0 / 3.7 / 3.1 / 2.2 / 2.2
If the WHO-Treatment did not work at all, then please explain, where the improvement from 7.5 to 4.0 in the placebogroup on day 1 did come from.
Please be assured, I am somewhat familiar with maths and statistics. If you evaluate the linear regression for day 2 to day 6, the decrease for placebo is 0.51 per day, for verum 0.44 per day only. Please let me have your interpretation of this.
All the results claimed statistically significant are significant only, if you start out with the groups being independent – what they never have been from the start on. Group #2 contains what is not in group #1. So the main criterion of two independent groups – both standing the chance of having the same properties – does not apply. This reduces the significance of the Nepal result like any other from p = 0.036 to p = 0.25.
If you feel, that this is a wrong idea, then please explain how the age of the children was affected by the drug (p = 0.03).
George, where are you?
George says:
Saturday 14 September 2013 at 16:52
(Firstly, keep in mind that since you don’t have the data for the later study you cannot conclude it is the same with the later one. )
Before I answer your questions
1. how much difference of linear regression between and controls and verum is generally acceptable to denote something significant? ( Why do you calculate from day 2 and not from day 1 by the way ?)
2. What do you mean “Group #2 contains what is not in group #1″ and how this makes the group not independent?
3. So the main criterion of two independent groups – both standing the chance of having the same properties – does not apply – can you elaborate ?
4.. how the age of the children was affected by the drug (p = 0.03). – I dont geet this – can you explain?
thanks
The 2003 study states, as a reason that the results of the three studies can be analysed together, that “all three studies followed the same basic study design, including similar entry criteria, treatment assignment, follow-up schedule, outcome measures and data analysis”.
Sorry, I was out of town for a few days….
Somehow the reply button does not show for George’s reply dated September 17, so I add this explication as a new answer.
<< … since you do not have the data you cannot conclude … <<
Nice way of Jacobs to hide the truth, isn't it? And consider Mojo's response.
(1) This does not apply here. The difference is the wrong way round. I told you, there are two phases of recovery, one initial jump – I think due to the start of the rehydration therapy – and the slow recovery from diarrhea.
(2) and (3): Evaluation of significance, as Jacobs (together with manby others) does it, assumes that both groups are independent from the beginning. But they were not, because assigning the patients to the groups leaves group #2 with what is not in group #1. If group number one has more 'self healers' than is its fair share, group #2 is short of them by the same number. In statistics, independence is defined as both groups having the chance of teh same distribution. This was violated in Jacobs' analysis.
(4) Did you read the study? Then the big difference in the chilldren's age should have caught your eye. This difference of age was just as significant as the recovery from diarrhea was. This is something of a proof, that significant diffeences can be achieved by the randomisation process.
Norbert Aust said:
The number of levels of comments is limited to five deep to keep them manageable by stopping the indentation going too far and the text becoming too narrow! You did the right thing – simply look back up the comments to find the Reply button.
Isn’t this the one where a true believer took two mutually contradictory insignificant results and combined them to produce one significant one, provided you ignore the fact that they are mutually contradictory?
Yes but this does not mean they are the same.
Anyhow, you are free to find errors and provide reasons if you want.
The justification for analysing them together was that the the methods they used were the same. if they are not the same, then the meta-analysis of them that you have cited is not valid.
The authors wanted to see if the verum group would recover faster than the placebo group and it did.
It is a simple design.
Regarding your calculations – sorry but you have to calculate from day one for the following reason.
Both groups were treated the same and you cannot conclude before the trial that how much of the effect is due to hydrotherapy..this is why you are conducting the experiment – if you knew the degree of the hydrotherapy effect you would not need to conduct the experiment. In the first place..
The age factor is your speculation – several studies whose results are not favorable for homeopathy – does not differ in terms of age grouping – randomization of groups- It is easy to find examples but I will provide if you cannot find.
The same thing regarding group independence – it is not always possible to achieve this kind of independence in medical research – this is the same for every protocol in placebo control trials. Isn’t that common practicec in designing a sound trial.?
(You know of course that trials do not include an untreated group, so determining the actual size of the placebo effect, compared to totally untreated patients, is indeed challenging .in every trial)
“Regarding your calculations – sorry but you have to calculate from day one for the following reason. ”
No, I do not. recovery clearly took place in two very distinctive phases – day 1 and day 2 to 6. And the children received two different medications: hydration and homeopathy/placebo. I am considering both phases seperately.
“Both groups …. not need to conduct the experiment”
Sorry, I do not understand this whole paragraph. Target was to find evidence for the positive effect of homeopathy, wasn’t it?
The ‘age factor’ is only alluded to for to show that ‘highly significant’ results could be produced by mere chance alone. I do not know if the age was of any consequence for the recovery.
I know it might be impossible to achieve true independence of the two groups – but this does not justify to utilize wrong maths. You can allow for this fact by using proper tests for significance (‘goodnbess of fit’-like tests).
“considering both phases separately ” is a wrong idea. The authors do NOT know the OVERALL effect of homeopathy ( if any ) during the entire 7th day period. They don’t know the mechanism.
They only thing they know is that at the last day the verum group performed better between 2 groups which have been separated by random methods including age as in it is done typically in placebo control trials. Look at at the table 2.
By the same mode for every trial you can say the same thing : ‘highly significant’ results could be produced by mere chance alone. This
In every placebo trial in conventional and homeopathic research, groups are separated the same way – and they are considered independent.
What is “wrong” with the math exactly ???
– can you show me an example form another placebo trial you think it is has been “correctly” designed ?
‘… is a wrong idea.’
Why? Plot the data and see, that there are two distinct phases.
‘… that at the last day …’
Wrong. In the Nicaragua trial the only point in time having significant difference was day 3 out of 5, not before and not after.
‘In every placebo trial in conventional and homeopathic research, groups are separated the same way – and they are considered independent.’
… which does not render this approach correct. I do not know if all the PCTs out there have the same problem – but the approach is wrong. The groups are not independent from the start on, at least not by a commonly accepted definition in statistics.
‘What is “wrong” with the math exactly???”
As I said: you should apply a single sample t-test or a goodness of fit test or any other test comparing verum to the pooled results of all the participating patients in total. This would give the probability to have this result by the allocation of patients alone, which would be the sole cause for inter-group variance in a placebo vs. placebo (= not effective homeopathic remedy) trial. For really efficacious remedies this raises the stakes just a little – maybe too high for not so very efficacious remedies.
The fact remains : there is significant difference measured by the same statistical tools used in placebo control trials in conventional medicine.
Typically placebo control trials use this protocol, design and statistical methods. They cannot really have independent groups the way you describe it.
iIf this design is “wrong” then all placebo control trials in medical research are useless and I agree with you.