Homeopathy trials can be positive — the reason why is bad news for homeopaths

    8 May 2016

    Whenever I write about the subject of homeopathy, I am sure to get a flurry of ad hominem attacks. They usually focus on well-rehearsed themes, calling me an unqualified fraud, an untrustworthy liar, an incompetent scientist and a corrupt recipient of bribes from ‘Big Pharma’.

    If you don’t believe me, please read some of the comments to my last post.

    All of these allegations have in common that they are untrue and potentially libellous, of course. In my view, they merely raise the suspicion that sizeable sections of homeopathy have deteriorated into a cult of irrational fanatics.

    Another comment voiced in response to my previous article is that, 26 years ago, I published a positive trial of homeopathy for treating varicose veins; this allegedly renders my criticisms of homeopathy inconsistent and arguably not credible (there even is a website that makes this claim). As the first part of this argument is correct, I feel the need to provide an explanation.

    I remember the trial in question well. We conducted it during my time in Vienna, and I am still proud of several innovative ideas that went into it (for instance, we developed an accurate method for measuring leg volume for it). Here is the abstract in full:

    ‘The aim of this study was to test the effectiveness of a combined homeopathic medication in primary varicosity. A well-defined population of 61 patients was randomised into active medication (Poikiven®) or placebo. Both were given for 24 days. At the start of the trial, after 12 days of medication and at the end of the study, objective and subjective parameters were recorded: venous filling time, leg volume, calf circumference, haemorheological measurements and patients’ symptoms such as cramps, itching, leg heaviness, pain during standing and the need to elevate the legs. The results show that venous filling time is changed by 44 per cent towards normal in the actively treated group. The average leg volume fell significantly more in this group, but calf circumferences did not change significantly and blood rheology was not altered in any relevant way. None of the patients reported side effects. Subjective complaints were relieved significantly more by Poikiven than by placebo. These results suggest that the oral treatment of primary varicosity using Poikiven is feasible.’

    So, there you have it: the most ‘notorious’ critic of homeopathy has shown that a homeopathic remedy is clearly superior to placebo in normalising important objective signs as well as subjective symptoms of varicose veins. Does that not render me inconsistent? Does it not undermine my credibility? Are homeopaths not entitled to be triumphant? And does that not mean that homeopathic remedies work after all?

    Disappointingly for homeopaths, the answer to these questions is no.

    To understand why, we need to look at the nature of homeopathic remedies. Most homeopathic remedies sold today are highly diluted — most but not all. The remedy we tested in our trial contained Aesculus D1 12,5 ml, Arnica D1 2,5 ml, Carduus marianus D1 5 ml, Hamamelis D1 10 ml, Lachesis D6 5 ml, Lycopodium D4 5 ml, Melilotus officinalis D1 10 ml. Take just the first of these ingredients, Aesculus (horse chestnut). This is a herbal medicine that has been well documented, even via a Cochrane review, to be effective for the symptoms of varicose veins. The fact that our remedy contained Aesculus in the D1 potency means that it was diluted merely by a factor of 1:10. So, for all intents and purposes, it was a herbal medicine, and it is utterly unsurprising that our study found it to be effective.

    Rather than proving homeopathy right, our study might even show the main principle underpinning homeopathy to be wrong. Homeopathy relies on the ‘like cures like’ principle. Therefore, if the assumptions of homeopathy were correct, a homeopathic dilution of Aesculus would have to worsen the symptoms of our patients rather than alleviating them. Seen from this perspective, our trial showed that the most fundamental axiom of homeopathy, the ‘like cures like’ principle, is incorrect.

    This story tells us that some trials of homeopathy are positive because they use medications which are homeopathic only by name. To make this perfectly clear: homeopathic remedies beyond a C12 potency contain nothing (these are the ones on sale in pharmacies and other outlets), less diluted remedies contain usually not enough to be pharmacologically active, and very low potencies, like the D1 potency used in our trial, are hardy diluted at all and thus contain substantial amounts of active ingredients.

    Homeopaths tend to profit from this confusion, for instance, when they rely on reviews of homeopathic trials which generate somewhat positive findings simply because they include studies of very low potency remedies. An example is the Cochrane review which found preliminary data in support of homeopathy as a treatment of adverse effects during cancer therapy. This result is due to remedies in very low dilutions. Homeopaths thrive on this complexity as it seems to negate the total uselessness of their remedies.

    So, are homeopathic remedies placebos or aren’t they? As we have seen, the answer is less straightforward than anticipated. The vast majority (99.9 per cent of those sold over the counter today, I estimate) are highly diluted and therefore contain either no or far too few active molecules to have any effect whatsoever. It is this vast majority that people refer to when they rightly insist that homeopathy is a placebo therapy. But we should not forget that, in addition, there are homeopathic remedies that are not highly diluted, like Aesculus D1 or Arsenicum D1. The former might help your varicose veins and the latter can kill your grandma.

    Edzard Ernst, emeritus professor at the University of Exeter, is the author of A Scientist in Wonderland and the awardee of the John Maddox Prize 2015 for standing up for science. He blogs at