Evidence: It’s whatever you want it to be

I have just sat through the an oral session for the evidence check on homeopathy, called by the House of Commons Science and Technology Committee. Well I wasn’t there in person, I watched it on the excellent video page. You can still see it if you are quick. Before I go on, aren’t we all glad they have reverted to a sensible name for this committee? Anyway, it was almost two hours very well spent. The transcript will be what endures for the public record, but that doesn’t capture all the body language and hesitation that attended so many of the answers to questions. I just want to pick out some bits that I think are worth our attention – in no particular order.

What astounded me more than anything was the staggering ignorance of Robert Wilson on matters of statistics. He was appearing for the British Association of Homeopathic Manufacturers, and was challenged on the evidence that homeopathy actually works. He launched a rambling discourse on sample size in clinical trials, stating categorically that most studies were too small (probably true), and that no study with less that 500 patients could be of any use. He seemed to be totally unaware of how sample sizes actually are calculated, and ignored the other factors such as expected differences, and variability. Only the number of patients mattered to him – one size fits all. This is not an esoteric subject, it is bedrock for anyone involved in life sciences. If he is the champion selected by the serried ranks of homeopathy companies, they can expect to lose their colours pretty quickly.

Wilson appeared in the first session, alongside representatives of the Royal Pharmaceutical Society of Great Britain (RPSGB), Sense About Science, and Boots, plus Dr Ben Goldacre of Bad Science fame. Everybody was asked whether they agreed that homeopathy works. Paul Bennett from Boots had to admit that there was no evidence, but regaled the committee with the mealy-mouthed flannel about customer choice that we have come to expect from his amoral employer. In summary, of the five witnesses, four said there was no evidence – and one of those was happy to carry on making money from gullible customers in the knowledge that they were being misled.

The second session was a bit more interesting. Dr Peter Fisher, director of the Royal London Homeopathic Hospital, was asked by the excellent Dr Evan Harris MP why the water used to prepare a remedy only retained the memory of that tincture and not everything else that had previously been dissolved in it. Fisher declared that the water was purified by double distillation, which removed all previous memories. Now this is an interesting hypothesis, and I am surprised that nobody picked him up on it. He is saying that double distillation resets the memory to zero, a claim that can readily be tested. We just need to prepare a standard remedy and divide it in half. One half we then double distil and use it to prepare a new remedy. We then give each remedy to patients in a clinical trial. There won’t be any difference of course, because neither group of patients will respond. If Fisher knows any other way of differentiating two remedies I’ll be avidly interested.

Another one that slipped through the committee’s net was a curious claim from Robert Mathie, research adviser to the British Homeopathic Association (BHA). He was reacting to criticism of high dilutions beyond Avogadro’s number, stating that many remedies were not so highly diluted and did have some remnants of solute. These had apparently been subject to tests, with positive results. Nobody seemed to remember the so-called `Law of Infinitesimals’, whereby potency increases with higher dilution. Is Mathie now repudiating this? But he got away with it.

One committee member was rather more sympathetic to homeopathy. Ian Stewart MP asked whether there was any evidence that homeopathy did not work, but nobody reacted to the essential silliness of the question. Had I been there, I would have said that I can’t prove that there is not a teapot orbiting the sun between Earth and Mars, but that there is no point in behaving as if there is (with apologies to Bertrand Russell). Such a question, from a member of the Science and Technology Committee, demonstrates a basis misunderstanding of science. The same member asked about evidence for harm from homeopathy, and nobody could provide any concrete examples. I would have mentioned Dr Marisa Viegas who in effect killed a heart failure patient by telling her to stop all drugs and take homeopathy. This was in the context of a question about regulation, and the committee should be told that the GMC only suspended Viegas for a year, and then struck her off because she still thought she was right. Sorry – I can’t give you a link for that case because it’s not on the GMC website (only goes back two years), but I have the minutes of the GMC hearing so email me if you would like it.

There was a lot of discussion about dangerous claims by homeopaths, particularly in relation to malaria prophylaxis. I would have liked to see Mathie pressed a lot harder on why no BHA member has been disciplined for making false claims. Bizarre belief systems always protect themselves from criticism by saying that they don’t support extremism, but they never actually keep their house in order by dealing with their own extremists. The result is of course more extremism.

In general, it surprised me that nobody wondered why we were having this debate about evidence at all. The homeopaths were stating that there is good evidence, and their opponents, led by the redoubtable Edzard Ernst, stated the opposite. We don’t have any such debate about the majority of orthodox drugs. It is usually as clear as the nose on your face whether a drug is effective or not, even though the impact of that on clinical practice might need more clarification. What we have with homeopathy is an elusive spectre of supposed efficacy which evaporates in the cold light of day.

Do look at the written submissions. Mine is there, but frankly the response from sceptics has been pretty pathetic – in terms of numbers not quality. Where were you all? The usual exponents of smokescreen science are there, notably the sociologist Clare Relton who loves splitting hairs about words. I need to steel myself to read Milgrom’s. Interestingly, the Prince’s Foundation takes pretty much the same line as Boots.

Well, I am looking forward to Monday’s session, with Professor Kent Woods of the MHRA, and Mike O’Brien MP and Professor David Harper from the Department of Health. How they defend what they have done with licensing will be an education. I don’t mean what they say (that will be the usual obfuscation), but their facial expressions will be a picture.