Bad Science. Ben Goldacre

line going down the middle, then the trial showed no clear difference either way. One last thing: the longer a horizontal line is, the less certain the outcome of the study was.

      Looking at the blobbogram, we can see that there are lots of not-very-certain studies, long horizontal lines, mostly touching the central vertical line of ‘no effect’; but they’re all a bit over to the left, so they all seem to suggest that steroids might be beneficial, even if each study itself is not statistically significant.

      The diamond at the bottom shows the pooled answer: that there is, in fact, very strong evidence indeed for steroids reducing the risk—by 30 to 50 per cent—of babies dying from the complications of immaturity. We should always remember the human cost of these abstract numbers: babies died unnecessarily because they were deprived of this life-saving treatment for a decade. They died, even when there was enough information available to know what would save them, because that information had not been synthesised together, and analysed systematically, in a meta-analysis.

      Back to homeopathy (you can see why I find it trivial now). A landmark meta-analysis was published recently in the Lancet. It was accompanied by an editorial titled: ‘The End of Homeopathy?’ Shang et al. did a very thorough meta-analysis of a vast number of homeopathy trials, and they found, overall, adding them all up, that homeopathy performs no better than placebo.

      The homeopaths were up in arms. If you mention this meta-analysis, they will try to tell you that it was a stitch-up. What Shang et al. did, essentially, like all the previous negative meta-analyses of homeopathy, was to exclude the poorer-quality trials from their analysis.

      Homeopaths like to pick out the trials that give them the answer that they want to hear, and ignore the rest, a practice called ‘cherry-picking’. But you can also cherry-pick your favourite meta-analyses, or misrepresent them. Shang et al. was only the latest in a long string of meta-analyses to show that homeopathy performs no better than placebo. What is truly amazing to me is that despite the negative results of these meta-analyses, homeopaths have continued—right to the top of the profession—to claim that these same meta-analyses support the use of homeopathy. They do this by quoting only the result for all trials included in each meta-analysis. This figure includes all of the poorer-quality trials. The most reliable figure, you now know, is for the restricted pool of the most ‘fair tests’, and when you look at those, homeopathy performs no better than placebo. If this fascinates you (and I would be very surprised), then I am currently producing a summary with some colleagues, and you will soon be able to find it online at badscience.net, in all its glorious detail, explaining the results of the various meta-analyses performed on homeopathy.

      Clinicians, pundits and researchers all like to say things like ‘There is a need for more research,’ because it sounds forward-thinking and open-minded. In fact that’s not always the case, and it’s a little-known fact that this very phrase has been effectively banned from the British Medical Journal for many years, on the grounds that it adds nothing: you may say what research is missing, on whom, how, measuring what, and why you want to do it, but the hand-waving, superficially open-minded call for ‘more research’ is meaningless and unhelpful.

      There have been over a hundred randomised placebo-controlled trials of homeopathy, and the time has come to stop. Homeopathy pills work no better than placebo pills, we know that much. But there is room for more interesting research. People do experience that homeopathy is positive for them, but the action is likely to be in the whole process of going to see a homeopath, of being listened to, having some kind of explanation for your symptoms, and all the other collateral benefits of old-fashioned, paternalistic, reassuring medicine. (Oh, and regression to the mean.)

      So we should measure that; and here is the final superb lesson in evidence-based medicine that homeopathy can teach us: sometimes you need to be imaginative about what kinds of research you do, compromise, and be driven by the questions that need answering, rather than the tools available to you.

      It is very common for researchers to research the things which interest them, in all areas of medicine; but they can be interested in quite different things from patients. One study actually thought to ask people with osteoarthritis of the knee what kind of research they wanted to be carried out, and the responses were fascinating: they wanted rigorous real-world evaluations of the benefits from physiotherapy and surgery, from educational and coping strategy interventions, and other pragmatic things. They didn’t want yet another trial comparing one pill with another, or with placebo.

      In the case of homeopathy, similarly, homeopaths want to believe that the power is in the pill, rather than in the whole process of going to visit a homeopath, having a chat and so on. It is crucially important to their professional identity. But I believe that going to see a homeopath is probably a helpful intervention, in some cases, for some people, even if the pills are just placebos. I think patients would agree, and I think it would be an interesting thing to measure. It would be easy, and you would do something called a pragmatic ‘waiting-list-controlled trial’.

      You take two hundred patients, say, all suitable for homeopathic treatment, currently in a GP clinic, and all willing to be referred on for homeopathy, then you split them randomly into two groups of one hundred. One group gets treated by a homeopath as normal, pills, consultation, smoke and voodoo, on top of whatever other treatment they are having, just like in the real world. The other group just sits on the waiting list. They get treatment as usual, whether that is ‘neglect’, ‘GP treatment’ or whatever, but no homeopathy. Then you measure outcomes, and compare who gets better the most.

      You could argue that it would be a trivial positive finding, and that it’s obvious the homeopathy group would do better; but it’s the only piece of research really waiting to be done. This is a ‘pragmatic trial’. The groups aren’t blinded, but they couldn’t possibly be in this kind of trial, and sometimes we have to accept compromises in experimental methodology. It would be a legitimate use of public money (or perhaps money from Boiron, the homeopathic pill company valued at $500 million), but there’s nothing to stop homeopaths from just cracking on and doing it for themselves: because despite the homeopaths’ fantasies, born out of a lack of knowledge, that research is difficult, magical and expensive, in fact such a trial would be very cheap to conduct.

      In fact, it’s not really money that’s missing from the alternative therapy research community, especially in Britain: it’s knowledge of evidence-based medicine, and expertise in how to do a trial. Their literature and debates drip with ignorance, and vitriolic anger at anyone who dares to appraise the trials. Their university courses, as far as they ever even dare to admit what they teach on them (it’s all suspiciously hidden away), seem to skirt around such explosive and threatening questions. I’ve suggested in various places, including at academic conferences, that the single thing that would most improve the quality of evidence in CAM would be funding for a simple, evidence-based medicine hotline, which anyone thinking about running a trial in their clinic could phone up and get advice on how to do it properly, to avoid wasting effort on an ‘unfair test’ that will rightly be regarded with contempt by all outsiders.

      In my pipe dream (I’m completely serious, if you’ve got the money) you’d need a handout, maybe a short course that people did to cover the basics, so they weren’t asking stupid questions, and phone support. In the meantime, if you’re a sensible homeopath and you want to do a GP-controlled trial, you could maybe try the badscience website forums, where there are people who might be able to give some pointers (among the childish fighters and trolls…).

      But would the homeopaths buy it? I think it would offend their sense of professionalism. You often see homeopaths trying to nuance their way through this tricky area, and they can’t quite make their minds up. Here, for example, is a Radio 4 interview, archived in full online, where Dr Elizabeth Thompson (consultant homeopathic physician, and honorary senior lecturer at the Department of Palliative Medicine at the University of Bristol) has a go.

      She starts off with some sensible stuff: homeopathy does work, but through non-specific effects, the cultural meaning of the process, the therapeutic relationship, it’s not about the pills, and so on. She practically comes out and says that homeopathy is all about