Download Lectures on Biostatistics (1971). Corrected and searchable version of Google books edition
Download review of Lectures on Biostatistics (THES, 1973).
CAM
Nobody knows the cost! But here is some information that I found by use of the Freedom of Information Act 2000…
Read full entry on the original IMPROBABLE SCIENCE page.
[This post was transferred from the old Improbable Science page]
It’s not just homeopathy. The Sunday Times, (26 February, 2006) reports that the National Health Service has fallen for another scam.
“IT COULD be called the Cleopatra Effect. Magnetic therapy, which has held the rich and powerful in thrall from ancient Egypt to modern Downing Street, is about to be made available on the National Health Service.
NHS accountants are so impressed by the cost-effectiveness of a “magnetic leg wrap” called 4UlcerCare that from Wednesday doctors will be allowed to prescribe it to patients.”
This is nicely timed to coincide with an Editorial in the British Medical Journal, by Finegold & Flamm (2006) (click to download pdf file). The editorial title was “Magnet therapy. Extraordinary claims, but no proved benefits“. They conclude
“Patients should be advised that magnet therapy has no proven benefits. If they insist on using a magnetic device, they could be advised to buy the cheapest – this will alleviate the pain in their wallet,”
For example, Carter et al,, 2002 found no detectable effect of magnet therapy for treatment of wrist pain Attributed to Carpal Tunnel Syndrome (30 patients,double blind, careful controls). Winemiller et al., 2003 (Journal of American Medical Association, 290, 1474–78), found no benefit of magnets vs sham-magnets in treatment of plantar heel pain in 101 patients.
Magnets are said to be one of Cherie Blair’s several curious and irrational beliefs. It is alleged, according to the Daily Telegraph, that “Cherie Blair did not allow her youngest child, Leo, to have the controversial MMR vaccine and instead asked a New Age healer to wave a “magic” pendulum over him”. A few more examples are documented here and here.
The accountants at the Prescription Pricing Authority have decided that the “the magnets will save money on bandages and nurses’ time by healing the wounds.” I dare say they could save even more money by removing all effective treatments.
The evidence in favour of the magnetic treatment all seems to come from a Dr Nyjon Eccles. The Sunday Times describes him as an “NHS GP in north London”, but elsewhere he is described as "Founder, CEO and Medical Director of the Chiron Clinic" in Harley Street. A look at their web site shows that they offer a full range of alternative scams. Cancer patients can get
“LYMPH DETOXIFICATION – This is achieved by non-invasive scalar, oxygen-fed light beam therapy. This helps to detoxify the tissues by assisting the body in dissolving lymph blockages and restoring normal lymph flow using the Nobel quantum scalar technology coupled with oxygen for remarkable healing potential.”
This is total gobbledygook, designed to take advantage of the desperate.
The only real evidence to be provided by Dr Eccles that the device works is a small (26 patients) double blind trial that has not yet been published in a peer-reviewed journal, and which suffered from a number of problems (dropouts, outliers). What, I wonder, does NICE think of evidence like this?
More on magnets and the PPA
In the discussion of magnets on the Badscience site, a Michael King says that 4ulcercare will be included in Part IX of the Drug Tariff because it meets the criteria of the Prescription Pricing Authority (PPA).
I presume this Michael King is Director of Planning and Corporate Affairs at the PPA, though he does not say so.
Michael King says
“There is no judgement offered about whether a product in the Drug Tariff
is more (or less) efficacious than any other, or the placebo effect.”
The criteria for inclusion in
Part IX of the Drug Tariff () include, in section 10 iii, “They are cost effective”
Will he please explain how a device can be cost-effective, if it is ineffective (relative to placebo)?
What the PPA says
Michael King has replied to my question by email (1 Mar 2006). He says
“The cost-effectiveness threshold for inclusion in the Drug Tariff is met if the ‘effectiveness’ of the device, as seen in data submitted by the manufacturer in support of the application, exceeds its cost to the NHS. ”
Sadly this is still ambiguous. It seems to suggest that that whatever data are submitted by the manufacturer are taken at face value, without any attempt to evaluate their quality. So I phoned King to ask if this was the case. He was helpful, but he said that it was not the case. He told me that the data were subject to some sort of low level evaluation, short of the sort of evaluation that NICE would do. This seems to contradict his earlier statement (above) that inclusion in the Tariff implies no judgement about whether a device is better than a placebo.
King said also that listing in the Tariff
“. . . is not a licensing decision nor a recommendation akin to the outcome of a NICE review”
The problem is, of course, that listing is seen as a recommendation by the public, by the Daily
Mail, and certainly by the manufacturer.
One thing, at least, is clear in this case. Whatever evaluation was done, it was done very badly. But in order to try to find out exactly what evaluation was done, and by whom, I’m having to resort to the Freedom of Information Act.
The results are here and here..
What NICE says
Fraser Woodward (Communications Manager, National Institute for Health and Clinical Excellence (NICE)) writes as follows.
“The test of “cost effectiveness” applied by the PPA when determining whether or not a device should go on the tariff is very different to the way cost effectiviness is assessed by NICE”
That is pretty obvious, but how is the public meant to know that, when they hear that the NHS has declared a treatment to be ‘cost-effective’, that statement can mean two entirely different things according to which part of the bureaucracy the statement comes from?
A good chance was missed to convey the facts and the science. Well below the BBC’s usual standard for science programmes.
Read full entry on the original IMPROBABLE SCIENCE page.
This entry has been transferred from the old IMPROBABLE SCIENCE page..
The Open University is a great institution. Its first vice-chancellor was Walter Laing Macdonald Perry . Before he took that job, he was professor of Pharmacology in Edinburgh (and one of my Ph.D. supervisors). He must be turning in his grave at the new OU course, K221 – Perspectives on Complementary and Alternative Medicine .
The course description sounds harmless enough, ” This course provides an accessible but rigorous introduction to complementary and alternative approaches to health.”. But just how rigorous is it? The game is given away when you see that the “experts” seem mostly to be true believers, people who make their living from alternative medicine. How can such people be expected to the merits of the systems of beliefs
that are the basis of their livelihood? It is rather like having a rigorous discussion about the existence of god in which all the course tutors are priests. Take some examples.
The bit about Testing Therapies is by Elaine Weatherley-Jones, She is in clinical practice as a homeopath. And as you might expect the three pages on the web about ‘testing therapies’ are highly partisan and selective. Try these quotations.
“In homeopathy, the vital force is said to be responsible for maintaining health, combating disease by recruiting the body’s natural tendency to cure itself. In the homeopathy model, disease occurs when the vital force is not working efficiently to keep the balance of health.”
“. . . in The Manual of Conventional Medicine for Alternative Practitioners : “The essence of alternative medical thought is that there is a vitalistic principle behind and encompassing any physical object”, explaining that “vitalistic” means that there are “objects which are non-physical in part or whole”. The vital
force of homeopathy and qi of TCM are non-physical – it’s impossible to see them, no matter how powerful an electron microscope was used. Qi and the vital force are ideas that are put forward to explain how the body heals itself.”
“Vital forces”? Which century are we living in? If this were offered as social studies, perhaps it would not matter, but the Open University is offering this course as part of a B.Sc. degree. Then we get the usual weasel words about the impossiblity of testing empirically whether alternative medicines (CAM) produce an effect, regardless of how they work. That is the important question. After all we are pretty vague about how some conventional drugs work. In a mind-boggling passage we are told that it is impossible to test CAM against a placebo, but quite possible to test CAM against an orthodox treatment.
It seems, incidentally, that the author’s grasp of pharmacology, and of the literature, is a bit weak. “. . . when Belon and his colleagues reported research in 2004 in the journal Inflammation Research , they showed that ultra-high dilutions of histamines (which are proteins involved in allergic reactions and causes, for example, inflammation of the breathing tubes in asthma) are active in influencing human cell activity. ”
Histamine (there is only one), is not a protein? This is meant to be a university course! Belon, of course, is a committed homeopath. This passage conveniently ignores the fact that his experiments have been repeated at least twice by respectable scientists, and they find no such effect. Surprisingly enough, they find that no drug gives no response. Amazingly, it seems that the OU would have us think otherwise.
The rest of the course seems to be much the same. The dispassionate expert on Herbal medicine is a herbal practitioner who makes his living from it, and is just as uncritical as one might expect in that circumstance. Unlike Weatherley–Jones, though, he does come clean (more or less) about the lack of evidence as to efficacy of herbal remedies (though that evidently does not deter him from practising the subject).
The “expert” on acupuncture , Rosey Grandage, is a bit more interesting on the history of her subject, but is every bit as committed to CAM as the others. She works at the University of Westminster as course leader of the Diploma in Qi Gong Tuina and also lectures on the BSc Acupuncture course. ” Rosey practices as a physiotherapist, acupuncturist and tuina practitioner in West London”. Hardly an unbiased observer. “ . . . it is this growing popularity which answers the question of whether acupuncture has a place in the modern world.”
Is it not obvious that the long persistence, and popularity, of an idea cannot possibly be used as an index of truth? One merely has to think of the long-persistent and popular ideas about the ‘one true god’. Clearly at most one of these can be true. The history of medicine is replete with popular and persistent ideas that turned out to be untrue. Take nux vomica . For hundreds of years conventional medicine regarded strychnine as a ‘tonic’. That persisted right up to the 1950s. But it became apparent that it just did not work, and strychnine, and the very word ‘tonic’, vanished from the vocabulary of rational medical people. You
won’t be surprised to find, though, that is still widely touted by fraudulent herbalists.
It could be argued that the course is intended as sociology rather than science, though the course description does not say so, and the course can count towards a BSc. Even as sociology though, it would seem better if the viewpoint of the tutors was rather broader.
The course books
I have now obtained copies of the three course books that were used for K221last year. They are indeed written largely as sociology not as science. But it is a highly biased sort of sociology, as one might have expected from the commitments of the authors. Although there are occasional references to lack of evidence, this does not seem to deter the authors from their relentless pursuit of the ‘integration’ of CAM into medical practice.
A superficial reading by a naive student might give an impression that the books are a “rigorous introduction to CAM”. The more sophisticated student is likely to see them as subtle, even insidious, propaganda. I won’t claim to have read all three books. Opening almost any page makes obvious their not-very-hidden agenda.
Here are a few examples from CAM: Structures and Safeguards (eds. Geraldine Lee–Treweek, Tom Heller, Hilary MacQueen, Julie Stone and Sue Spurr).
Chapter 5 (Homeopathy: principles practice and controversies) contains a ludicrously biased account ot the affair of Jacques Benveniste (see here and here ). There is no mention of the fact that his results were disproved at the time, and at least twice since, I know of only one group that has claimed similar results, and that group, like Benveniste’s, contained committed homeopaths. There is no mention of Beneveniste’s two Ignobel prizes. There is no mention of the fact that after he left France in disgrace, he went on to claim that the properties of the alleged memory of water could be sent by email, a claim so absurd that it has not persisted even within CAM.
This chapter uses the standard CAM trick of redefining the word efficacy. Rather than its usual meaning of having an effect greater than placebo, it is conveniently redefined to mean, roughly, ‘patients say they feel better’,
“Although the issue of whether or how homeopathic remedies ‘work’ is a major bone of contention for medical scientists, the fact that they perceive that the remedies do work makes it attractive to many orthodox medical practitioners.”
So that’s OK then. Don’t trouble yourself with what’s true.
Chapter 1 of ‘CAM: Structures and Safeguards’ has the title “Knowledge, names, fraud and trust”, and is by Geralidine Lee-Treweek. It is a fine example of relativism -almost post-modernist in style. This is a discussion of knowledge in which the words ‘true’ and ‘false’ barely appear. The student who lent me
the book has scribbled in the margin “If it is not true and right –then it is not knowledge”. Pass the student, fail the tutor . (in fairness, it has to be pointed out that the student passed with distinction, despite her scepticism.)
So does Open University course K221 really give you a “rigorous introduction to complementary and alternative approaches to health. ” No it seems that it does not. Here are some more reasons.
A student who has successfully completed course K221 has told me that
- “It was very anti-science and anti-orthodoxy in places”
- “I had several ‘discussions’ with her [OU tutor] in our online tutor group, some about her anti-vaccination stance and the fact that she was happy to give homeopathic anti-malarials to travellers.”
The argument that homeopaths at least do no harm ( see above ) seems to be destroyed by their advocacy of policies that will lead to more children getting measles, and which will contribute to the spread of malaria. There is an ultimate irony in OU tutors preaching against vaccination. Walter Perry, the first vice-chancellor of the OU, before he was my supervisor in Edinburgh, had been Director of Biological Standards at the Medical Research Council’s labs. In that job he had responsibility for introduction of polio vaccine in UK. That effectively eliminated the scourge of polio.
This is not what a real university should be doing, as part of a B.Sc. degree.
An article on the death of homeopathy, There has been long enough to get evidence, but it is not there.
Read full entry on the original IMPROBABLE SCIENCE page.
On 21 November, 2005, Dr David Spence appeared on the BBC’s Today Programme. He was being interviewed about a report that, he said, provided evidence for the effectiveness of homeopathy. In fact it does nothing of the sort.
Dr Spence’s paper was published in the Journal of Alternative and Complementary Medicine. It is not really research at all. They simply asked 6544 patients who had had homeopathic treatment whether they felt better or not. Half the patients (50.7%) said they were ‘better’ ot ‘much better’. A further 20% said they were ‘slightly better’. The patients who had homeopathic treatment were not compared with anything whatsoever!
This is reported in a straighforward way. What is quite ludicrous is the stated conclusion of the paper:
“The study results show that homeopathic treatment is a valuable intervention”.
It is obvious that there is not the slightest reason to attribute the answers given by patients to the fact that they had been given homeopathic treatment. That would be the crudest form of post hoc ergo propter hoc error. It does not even show that the homeopathic treatment was producing a placebo effect.
Papers like this do not add to human knowledge, they detract from it. By reverting to pre-enlightment forms of argument, they mislead rather than enlighten. To make matters worse, this work was done at public expense, by the Directorate of Homeopathic Medicine, United Bristol Healthcare, National Health Service Trust, Bristol, United Kingdom.
What on earth is a respectable hospital and medical school, like those in Bristol, wasting money with this sort of mediaeval hindrance to medical knowledge? We are truly living in an age of delusions.
Download the paper and see for yourself [ Spence DS, Thompson EA, Barron SJ. J Altern Complement Med. 2005, 11, 793-8. pdf file, 74 kb].
The Daily Telegraph recently published two reports about acupuncture. One said it worked. The other said it didn’t work. Needle cure effect ‘is not all in the mind’ By Catriona Davies, starts
Acupuncture has a genuine ability to relieve pain, scientists have found.
The other report was
Doubt cast on needle therapy for migraine By Nic Fleming, Health Correspondent, starts
Acupuncture is no better at reducing migraines than fake treatment, researchers say today.
A study involving more than 300 patients found the healing method did reduce headaches, but only by the same amount as placing needles at non-acupuncture points.
Unfortunately the Daily Telegraph‘s reporters did nothing to help the confused reader. No comment was made on the apparent contradiction. In this particular case, there is an obvious explanation.
The first (favourable) article was said to be published in Nature, though in fact it was published in Neuroimage (Pariente J., White P., Frackowiak , Richard S. J. & Lewith G. Neuroimage, 25. 1161 – 1167, 2005). Presumably the reporter had picked it up from a rather uncritical synopsis on the news site, nature.com. It was conducted on 14 patients with painful osteoarthritis. Contrary to the first line of the Telegraph‘s report (“Acupuncture has a genuine ability to relieve pain”), the work did not measure pain at all. In fact the summary of the paper says
“The three interventions, all of which were sub-optimal acupuncture treatment, did not modify the patient’s pain.”
What the paper actually did was to use positron emission tomography (PET) to measure ‘activation’ of certain parts of the brain when needles were inserted. Some parts were activated more by having acupuncture needles piercing ths body than by ‘stage-dagger’ needles which retracted and did not pierce the body. I have no criticism of these findings: my purpose here is to explore the apparent contradiction between this trial and another.
The second, unfavourable, report was of a much bigger study, 302 patients with migraine headaches. It was published by Linde et al. (Journal of the American Medical Association. 2005 293(17):2118-25). This study concluded
“Acupuncture was no more effective than sham acupuncture in reducing migraine headaches although both interventions were more effective than a waiting list control. ”
These two studies were on quite different conditions, used different methods, and very different numbers of patients. But suppose we take them at face value, are they not contradictory? No, not necessarily, because they used quite different sorts of control group.
The study that was reported as showing that acupuncture worked compared patients that had real acupuncture with patients who had treatment with “stage dagger” needles that appeared to pierce the skin but did not.
The study that was reported as showing that acupuncture did not work (Linde et al.) used a different sort of control group, acupuncture needles that pierced the skin but were inserted in the wrong places (as defined by the ‘principles’ of acupuncturists). No difference was found between ‘real acupuncture’ and control.
There is nothing incompatible about these two studies if one adopts the view that piercing the skin with a needle can produce a physiological response that makes the patient feel that other sorts of pain are less painful, but that it does not matter where the skin is pierced. The latter hypothesis means, of course, that all talk about “meridians”` and “energy flow” that acupuncturists use, is no more than mumbo jumbo.
Dr Lewith is quoted as saying, of the negative study, “This is a badly conceived study that just adds more confusion to the debate because it uses non-site specific acupuncture as a control.” Quite on the contrary, the Linde study seems more interesting to me, because it unveils the mumbo jumbo of meridians (at least for the particular points used in this study). Dr Lewith may find this conclusion unpalatable, but it is the obvious implication of this pair of studies.
Another study confirms that the principles of acupuncture are nonsense
A trial by Melchart and colleagues on 270 patients shows conclusively that acupuncture can indeed produce amelioration of tension-type headache when compared with no treatment, BUT the relief is produced whether or not needles are inserted at ‘acupuncture points’. Very similar results were found with ‘superficial needling at non-acupuncture points’. British Medical Journal, 15 August 2005. Yet again it is shown that the mumbo-jumbo of meridians and magic points is nonsense. Yet again, that is a good reason why universities cannot be expected to train acupuncturists. Our business is to minimise mumbo-jumbo, not to propagate it.
This letter to The Times points out the folly of making regulations that do not require any demonstration that the product works. Can you imagine a regulation for television sets that required only that they do no harm, but did not specify that they should show a picture?
| From Professor D. Colquhoun, FRS Sir, Congratulations on your report on the deficiencies found in complementary medicine practitioners (Body & Soul, January 10). In the face of such evidence it is natural to ask for more effective regulation of this very profitable industry. But the question is quis custodiet ipsos custodes? This question has serious implications for the universities as well as for the public (and the industry).The House of Lords report and the Government’s response to it, pointed out that the first step was to find out whether the complementary treatment worked (better than a placebo). They recommended that the Department of Health should fund research on complementary medicine, the first priority being to find out whether each therapy worked. The problem is that you cannot regulate properly an area when it is not, in most cases, known whether the product being offered has no effect above that of wishful thinking. This raises a serious question for universities, because it leads, naturally enough, to demands for better training. But how can a university run a course on a subject about which there is so little hard evidence? Tragically (for their own reputation), some of the new universities are running three-year BSc courses in such subjects as complementary therapies. I’m quite happy to believe that nice smells produce good placebo effects, but aromatherapy is not, by any stretch of the imagination, science, and in my view it is not honest to award Bachelor of Science degrees in it.The effect of such courses will be not to promote better regulation, but to give spurious respectability to an industry that, according to the Government, should (but does not) have, as its first priority, to find out what works and what doesn’t. Yours faithfully, DAVID COLQUHOUN, A. J. Clark Professor of Pharmacology, |
This is the story of my first incursion in to the fantasy world of alternative medicine.
I was asked by the producer of a television programme (QED) to look at a paper that claimed a beneficial effect of homeopathic treatment of fibrositis (Fisher, P., Greenwood, A., Huskisson, E. C., Turner, P., & Belon, P. (1989). Effect of homoeopathic treatment on fibrositis (primary fibromyalgia) British Medical Journal 299, 365-366.) [download pdf].
The homeopath, Peter Fisher, was kind enough to give me the raw data for re-analysis. Curiously. the two medical co-authors (apparently guest authors), neither of them homeopath, were reluctant to hand over the raw data.
It appeared from the paper that the crossover trial had been analysed incorrectly (each patient had been counted twice). When the results were analysed correctly, no significant effects were found.
Astonishingly, the British Medical Journal declined to publish the correction, but their rival, the Lancet, did so with alacrity (Colquhoun, D. (1990). Reanalysis of a clinical trial of a homoeopathic treatment of fibrositis. Lancet 336, 441-442.).[ download pdf ].
Incidentally, the result of this exercise, despite the fact that it had been commissioned by the television producer, was entirely misrepresented in the final TV programme. The producer was evidently less interested in discovering the truth, than in giving the public what he thought they wanted, i.e. wishful thinking. In this he must have been successful, because the first letter that I received after the programme was from a lady in Fulham, who asked me to recommend a source of homeopathic flu jabs for her cat.
It’s interesting, but not surprising that this correction has been universally ignored by advocates of homeopathy. Whether this is incompetence or dishonesty is impossible to say.
Both the House of Lords report on Complementary and Alternative Medicine, and the Government’s response to it, state clearly “. . . we recommend that three important questions should be addressed in the following order:”. (1) does the treatment offer therapeutic benefits greater than placebo? (2) the treatment safe? (3) how does it compare, in medical outcome and cost-effectiveness, with other forms of treatment?
These recommendations seem admirable, but they have not been followed. The money has gone, almost (if not completely) to projects that address the second and third questions, before it has been established that the treatments have anything other than a placebo effect. This interesting case is debated in Debate: UK government funds CAM research (Focus on Alternative and Complementary Therapies , 8, 397-401 (DC’s bit, pp 400-401)).
[Get PDF of whole debate]
|
Both the House of Lords report and the Government response to it, state clearly “… we recommend that three important questions should be addressed in the following order: . . .
These aims seem admirable, but to what extent do the projects that have been funded match these recommendations? The answer, sad to say, is that they do not seem to follow the recommended order of priorities at all. None of the studies in the first initiative (Tovey, Corner and Shaw) appears to address the question that the recommendations specify should be done first. With the possible exception of White, none of those in the second phase (Shaw, Barry, Weatherley-}ones, White and MacPherson) do. (Dr White, like most other recipients, has declined to provide any information about his project so it is impossible to be sure.) None really addresses ,the second priority directly. The third recommendation is worded much more vaguely than the first two, but it is The rationalisation given by some of the applicants for uncontrolled, or ‘pragmatic’ trials is that they are conducted under real clinical conditions and tell you what the patient actually thinks. It is quite true that, from the point of view of the patient, it does not matter in the least whether they feel better because of a placebo effect or because of a specific effect of the treatment. That is an important consideration but it is not the only one. If the first priority had been addressed first (which it has not) it is quite possible that the outcome could be that the entire effect could be a placebo effect. Such a possibility has been envisaged by no less a luminary of the CAM world than Peter Fisher (ref 1). If that were to turn out to be the case it might matter little to the patient but it would matter a great deal to universities, which are under continual pressure from CAM people to run degree courses (though only a few have acquiesced). If the whole effect were placebo, it follows that the ‘principles’ of homoeopathy, reflexology, etc. are mere mambo jumbo and so not appropriate for teaching in universities (or, indeed, anywhere else). The question of courses and ‘training’ cannot be considered until the first question is answered because, until then, we do not know if there is anything real to train people about. That is why it is the first priority. There would however, be a dilemma for clinical practice. The placebo effect does appear to be useful, so the question would then become how best to produce a good. placebo effect without too much intellectual dishonesty. Perhaps that is a question that deserves more research. The fact of the matter is that the Department of Health has ended up spending £1.3 million of public money in a way that directly contravenes the recommendations of the House of Lords and of the government (with one possible exception). They claim that this happened because very few applications were received that addressed the government’s first priority. That alone says something about the extent to which the CAM world is interested in tests against placebo hardly surprising since a negative result would destroy their livelihood. But, arguably, if few applications were received that addressed the first priority, then the funding should have been postponed until appropriate applications were forthcoming. The reason that this did not happen is, I fear, only too obvious. The judging panel was dominated by CAM people who clearly share the lack of interest shown by the rest of the CAM community in answering the most important question first. If such research must be done, because of public demand for it, it should have been organised by the Medical Research Council using the same criteria they would use for any other treatment. References 1 Fisher P, Scott DL. A randomized controlled trial |