Thursday, 27 September 2007

Dana Ullman MPH gives me some homework

Dana Ullman MPH (the only person I've ever seen put their qualifications on their blogger username) commented on my piece on Alan Bennett's description of alternative therapies here, suggested that I'm confused about what homeopathy is, and gave me some homework to do. I've read Ullman's webpage, and I can tell you that it's a farrago of nonsense, written in superficially scientific language.

I was aware of some of Dana Ullman's work previously. For example, there's an entire thread dedicated to him at JREF, which you can find here. He posted there (under the username James Gully) to write some utter nonsense about famous scientific figures who supposedly used or supported homeopathy. Having had his arse handed to him by several posters on the forum, he accused them of 'intellectual dishonesty' and disappeared. But I hadn't seen this webpage before. It's entitled 'Why homeopathy makes sense and works', but fails to demonstrate either. You can find it here.

Ullman starts off by talking about side effects. He writes "It should be noted that people often incorrectly assume that conventional drugs have 'side effects.' Actually, in purely pharmacological terms, drugs do NOT have side effects; drugs only have 'effects,' and physicians arbitrarily differentiate between those effects that they like as the effects of the drug, while they call those symptoms that they don’t like 'side effects.' This is akin to saying that the effects of a bomb are that it destroys buildings, but its side effects are that it kills people. Needless to say, one cannot truly separate out one effect from the other. The reason that drugs create 'side effects' that are often worse than the original disease is that these drugs tend to suppress the symptoms the sick person is experiencing and push them deeper into the person’s body."

In fact, the distinction between effects and side effects is simply that the effects are desired, and the 'side effects' are not. It's not an arbitrary distinction, and no-one is trying to claim that side-effects are not effects caused by the drug. That's why medicines are tested for safety before they are licensed for sale. The reason why drugs have side effects is that they contain biologically active substances. A good explanation for the lack of side effects with homeopathic remedies is that they do not contain biologically active substances, that is, they don't work.

Ullman describes the principles behind homeopathy. In a section headed 'Determining what a medicine can cure', he writes about homeopathic provings. In a proving, subjects are given a dose of a substance, and their 'symptoms' are recorded. It is then assumed that a small (or nonexistent) dose of the substance will cure the same symptoms. Ullmann calls these provings 'toxicological studies', but it's easy to see that as scientific studies they leave a lot to be desired. It's an exercise in the post hoc ergo propter hoc fallacy, in that it's impossible to know whether the 'symptoms' were caused by the dose because there are no adequate controls.

Ullman then goes on to describe how homeopathic remedies are made, and talks about the power of 'nanodoses'. Firstly, this is a little disingenuous, as it suggests that homeopathic remedies contain very small amounts of an active ingredient, whereas in many cases they contain no active ingredient at all. Perhaps 'nonodose' would be a better term. Ullman describes the process of dilution and 'succussion' (shaking) well enough, but then bizarrely states that "It is inaccurate to say that homeopathic medicines are extremely diluted; they are extremely 'potentized'". It is not inaccurate to say that the medicines are extremely diluted: they are extremely diluted, as Ullman shows when he talks about serial 1:10 or 1:100 dilutions being conducted up to 1,000,000 times (as an aside, how can this easily be done? If I assume one 'potentisation' step can be done in one minute, it would take nearly two years to do 1,000,000 times, assuming I work 24 hours a day). The homeopathic theory is that this dilution and shaking makes the remedy more 'potent'. Apart from being against common sense (which after all can be wrong), this also goes against the dose-response effect well known from pharmacology, i.e. that a greater dose causes a greater effect. In a section headed "Other evidence on the power of nanodoses" Ullmann writes about certain compounds that have biological effects at very low concentrations, or that have very different effects at low concentrations than they do at high concentrations. This is not relevant to homeopathy, where substances are supposed to be biologically active at zero concentration.

Ullman also writes about clinical evidence for homeopathy. One thing about this section is that Ullman seems to misunderstand p values, when he writes that p=0.008 "means that there was a 99.2% chance that this treatment was effective". It means that if you conducted the experiment 1,000 times, you would expect to get a positive result 8 times through chance. It doesn't tell you about biases, poor experimental design, or other problems with the study (there's a useful discussion of some of these things here, here and here). For any of the examples Ullman gives of studies showing benefits for homeopathy, there are several that show the opposite. A recent well-conducted meta-analysis in the Lancet looked at homeopathy versus 'conventional' treatments. It found that the best conducted studies showed no benefit for homeopathic remedies beyond placebo, whereas the conventional treatments did show a benefit beyond placebo. Tellingly, Ullman mentions New Scientist, a popular science magazine that does not publish original research.

In the last paragraph of the 'clinical evidence' section, Ullman mentions 'water memory'. As regular readers of hawk/handsaw will know (hi to all two of you!), a recent issue of the journal Homeopathy was dedicated to this concept. None of the papers in it showed any 'memory' effect relevant to homeopathy, as discussed here, here, here, here, here and here. It's wishful thinking, at best.

For me, there's nothing on Dana Ullman's page that makes me think that homeopathy 'makes sense and works'. Not only that, but I didn't learn anything I didn't already know about homeopathy.

Edit: Here's a link to the Respectful Insolence blog on the COPD study mentioned by Dana Ullman in the comments to this piece.

Tuesday, 25 September 2007

Alan Bennett and homeopathy

I bow to no-one in my admiration for the writing of Alan Bennett. It probably helps that he's a Yorkshireman, albeit from the evil that is Leeds. I've recently been re-reading his book 'Untold Stories'. In a piece entitled 'An average rock bun', Bennett writes about his experience with cancer of the colon, the approximate size of the tumour lending the piece its title. This is typically modest, unassuming and moving stuff, but what I found most interesting was Bennett's experience with homeopathy.

After an operation to remove the tumour, Bennett is due to begin precautionary chemotherapy. At the same time, he explores some alternative treatments, and after talking to his GP, decides that homeopathy might help. I'm somewhat ambivalent about this. After all, as long as Bennett continues with the chemotherapy, a homeopathic treatment can't do any harm, and even if it only has a placebo effect it might still help him to feel better. What is disgraceful is what happens when Bennett visits a "reputable complementary health clinic in Harley Street". He manages to get an appointment at 7am, and sees the complementary therapist. He explains that he has already decided on a course of chemotherapy, and wants to take homeopathic treatments alongside it.

As Bennett writes: "He proceeded to pour scorn on chemotherapy, the benefits of which he said were unproven, and when I didn't budge, rather sulkily conducted some finger-tip tests, which I took to be to do with the homeopathy, but done in such a perfunctory fashion I'm not sure he thought much of the point of this either. People kept coming in with whom he chatted, and throughout treated the business so casually and with such a disregard for my predicament, and presumed agony of mind, that it reminded me of the arrogant and unconcerned conventional doctors one used to come across thirty years ago".

Not very nice, is it? But that's not all. Bennett agreed to have some blood tests done by the complementary clinic. He then received a letter from the therapist, telling him that his blood test results were 'a lot worse than we thought' and that 'without sorting this out, your chances are much less than 50-50'. The cure was a course of vitamin injections. As a playwrite, writer and sometime art historian, Bennett is probably not up to speed with the latest medical research, but he's certainly no fool. As he writes, "All I saw was a barefaced attempt at medical blackmail and a doctor trying to panic me into using his clinic's doubtless expensive facilities". Bennett did not go back to the clinic.

So here we have an example of a homeopathic practitioner rubbishing proven therapies, and suggesting that they be replaced with homeopathy, and then playing on the fears of the patient to get them to take an expensive course of vitamin injections. This is extremely irresponsible and dangerous. Having complementary therapies used beside conventional treatments, under the supervision of a GP, is one thing. Advocating vitamin injections and magic water as a replacement for chemotherapy is a very different thing. Regardless of the science, complementary and alternative medicine needs to get its ethical house in order as a matter of urgency. Then again, the first question they would have to resolve is how you can ethically sell bullshit to people, so you can see why this would be a difficult enterprise.

Edit: If anyone knows what this is doing translated into German here, I'd be interested.

Wednesday, 19 September 2007

Even more on memory of water

I had a go at critiquing the recent Homeopathy special issue on the 'memory of water'. Ben Goldacre has now suggested that people who have been critiquing the issue could submit comments for publication to the journal. This is the way a civilised scientific debate is conducted, so I decided to give it a go. The following has been submitted as a letter to the editor (cheers to JJM and Rolfe at the JREF forums who proof-read it for me) :

Martin F. Chaplin presents an interesting overview on the structure of water1. Disappointingly, though, it seems to contain no useful evidence of a ‘water memory’ effect that would be relevant to the efficacy or otherwise of homeopathic treatments. As is well known, the probability of any dilution beyond about 12c (a dilution factor of 1x1024) containing a single molecule of the ‘mother tincture’ used to prepare the remedy is very low. So homeopaths require that water (or water/ethanol mixtures) somehow structurally ‘remember’ the mother tincture, and it is this structure that is responsible for any effect of homeopathic preparations. But Chaplin1 appears to be talking about an entirely different effect when he states that “If there is evidence that the history of a sample of water affects its properties, then the ‘memory of water’ concept is proven without the need for a rationale for its action” (p. 146). This apparently broad interpretation explains some of the examples given as “Evidence for the memory of water”, which otherwise appear to have little to do with homeopathic remedies, where the mother tincture must continue to influence water structure even when absent.

For example, Chaplin1 (p. 146) states that “human taste is quite capable of telling the difference between two glasses of water, processed in different ways (e.g. one fresh and one undrunk for several days)”. There is nothing mysterious about this. Depending on the circumstances, this can be due to outgassing of Cl2 and/or absorption of atmospheric CO2. This has nothing to do with how a mother tincture diluted out of existence has any effect on water structure. In the homeopathic context, what would be really interesting is if it were possible to taste the difference, say, between two 30c dilutions made from different mother tinctures prepared under identical conditions. Chaplin also refers to the Vybíral and Voráček paper in the special issue of Homeopathy2, stating that the authors “have shown that water changes its properties with time and its previous history”1 (p. 146). There is no doubt that this is an interesting paper, but the authors specifically conclude that their results are a consequence of ions dissolved in the water, as the effect they observe is not present when deionised water is used. Again, it is unclear how this is relevant to cases where mother tinctures are diluted out of existence. What Chaplin seems to be talking about in his paper is how impurities might affect water structure, as illustrated when he states “The water used for dilution is not pure relative to the putative concentration of the ‘active’ ingredient, with even the purest water considered grossly contaminated compared with the theoretical homeopathic dilution levels. This contamination may well have a major influence, and itself be influenced by the structuring in the water it encounters”1 (p. 148). Since the concentration of the mother tincture will always be dwarfed by the concentration of impurities, it is difficult to see why the mother tincture should have an effect more important than the impurities on the water structure. Homeopathic remedies are not ‘just water’: they will contain significant amounts of impurities, which might create interesting structures in the water. But how this would make them significantly different from any sample of water is not clear.

I am in agreement with Chaplin when he states that “simply proving that water does have a memory does not prove that homeopathic medicines work”. The best evidence as to whether homeopathic medicines work or not is to be found in randomised placebo-controlled trials of the medicines. A recent meta-analysis3 has demonstrated that the best-conducted trials show no effect for homeopathic remedies beyond placebo. The structure of water is certainly a fascinating subject, as Chaplin shows, but all the best evidence shows that homeopathy has no effect that requires an explanation.


  1. Chaplin, M.F. The memory of water: an overview. Homeopathy 2007; 96: 143-150
  1. Vybíral, B. and Voráček, P. Long term structural effects in water: autothixotropy of water and its hysteresis. Homeopathy 2007; 96: 183–188.
  1. Shang, A., Huwiler-Müntener, K., Nartey, L., Jüni, P., Dörig, S., Sterne, J.A.C., Pewsner, D., Egger, M. Are the clinical effects of homoeopathy placebo effects? Comparitive study of placebo-controlled trials of homoeopathy and allopathy. Lancet 2005; 366: 726-732.

Thursday, 13 September 2007

Patrick Holford seems to think I'm important enough to hassle

Very, very strange.

Patrick Holford is a nutritionist. In the UK, the title 'Dietitian' is protected, but anyone can call themselves a nutritionist. So I'm a nutritionist too. Holford's ideas are not without controversy, and he even has a debunking website, Holfordwatch, all to himself. He gets 6 canards at the excellent Quackometer website. Recently, Holford has been appointed a visiting professor at the University of Teesside. This was not something you could expect Prof. David Colquhoun to take lying down, and so he sent an FOI request to Teesside, asking for the documents, including Holford's CV, that were used to make the appointment.

Prof. Colquhoun found a number of anomalies on the CV, as did folk at Holfordwatch. I left a couple of comments at Prof. Colquhoun's website, after I searched for and failed to find some of the media endorsements listed in Holford's CV. Apparently, Holford has now contacted various people who also commented on Prof. Colquhoun's website, and asked them to stop posting there to agree with Prof. Colquhoun. It seems that he was trying to threaten anyone who had posted to the site. Holford apparently couldn't be bothered to find my e-mail address, instead asking other people that he had contacted to forward the e-mail to me and other commenters. This is pretty bizarre behaviour in itself, with Holford acting as if people posting on someone else's blog have some special way of knowing who each other are. In any case, as yet, I've received no communication from Holford himself.

So it seems that Patrick Holford thinks it worth hassling people posting comments on a blog. Why, I don't know, but there it is. Prof. Colquhoun himself refuses to be intimidated and has backed up his points here. As for me, I'm flattered that Holford thinks me important enough to bother with. I don't have even the level of microfame that something like the Holfordwatch blog has, and it must be doubted that any comment I make anywhere is going to have much of an effect on the Holford nutritional supplement empire.

Wednesday, 12 September 2007

New blog: Structure and Tectonics of Basins

I've decided to set up a separate blog, where I can discuss current research in the fields of structure and tectonics of basins and petroleum geology. You can find it here.

Friday, 7 September 2007

Denouncing your own publications as 'crap'

What happens when you have an academic system that prioritises quantity of publications over everything else? What happens when drug companies need eminent scientists to conclude that their stuff works? Predictably, you get perverse results. Like this one, detailed here and here (and all over the interweb).

In this case, it seems as if the putative author of the ghost-written work is happy to disown it. I can imagine similar cases where the author would be happy to put the resulting bogus publication on their CV, because it's nice to have a career. It's a vortex of shit, folks.