The Evidence

Rational Veterinary Medicine:

Home Papers Media RVM articles Contact

On Sale now!

Papers, listed by lead author: T-U

T

Taylor, S.M., Mallon, T.R. and Green, W.P., (1989) Efficacy of a homoeopathic prophylaxis against experimental infection of calves by the bovine lungworm Dictyocaulus viviparus. Veterinary Record Vol. 124 no. 1 pp. 15-7

Links: [abstract - vet rec/BMJ]

“Two groups of parasite-free calves, one of which had been treated with four doses of a homoeopathic oral vaccine for parasitic bronchitis due to Dictyocaulus viviparus and the other with a placebo, were infected at the rate of 25 infective larvae/kg bodyweight 18 days after the final dose. Both groups became severely affected by parasitic bronchitis, with clinical signs starting 13 days after infection. There were no discernible differences between the treated and control groups in their manifestations of resistance to D viviparus or their clinical responses to the disease produced.”

Yes, that’s right, homeopathy kills as many calves as a placebo - one hopes such a trial wouldn’t get past the ethics board these days.

Taylor, M.A., Reilly, D., Llewellyn-Jones, R.H., McSharry, C. and Aitchison, T., (2000) Randomised controlled trial of homeopathy versus placebo in perennial allergic rhinitis with overview of four trial series British Medical Journal Vol. 321 pp. 471- 476 [with appended comment by: Lancaster, T., Vickers, A.]

Links: [abstract pubmed]:[full text pubmed central]:[full text html BMJ]:[full text pdf BMJ]

Responses: [commentary - Vickers and Lancaster BMJ 2000 (go to the bottom of the page)]:[rapid responses BMJ] -

[Dawkins "flawed experimental design" BMJ]:[Miller "flawed statistics" BMJ]:[Windeler "infact it isn't positive" BMJ]:[Brown "was it really allergic rhinitis?" BMJ]:[Hadjicostas “Does even the wrong homeopathy work?”]

[apgaylard blog entry 2009]:[JREF thread]

“CONCLUSION: The objective results reinforce earlier evidence that homoeopathic dilutions differ from placebo.”

... or do they?

This study was the last of a series of four on the treatment of perennial rhinitis, or hay fever. They were carried out by largely the same authors in the period from 1985 to 2000, the others being Reilly et al 1985 in the British Homeopathic Journal; and Reilly et al 1986 and Reilly et al 1994 both in the Lancet. As with the first three, the authors considered the results from the 2000 paper confirmed homeopathy was different from placebo, and when they combined the results from all four trials they felt this gave further confirmation still. All four papers are looked at together here.

Although this set of works is considered one of the darlings of homeopathic literature - three of them having achieved the holy grail of being published in mainstream journals - they really are a pretty ineffective demonstration of what its proponents claim homeopathy has to offer.

There are many criticisms which can be made of them, some of the most effective appearing in the James Randi Educational Forum thread, but a considerable number appearing in the BMJ itself in the form of Rapid Responses - a sort of email equivalent to letters to the editor (all links above).

One of the most compelling and fundamental problems with the 2000 paper is described by Miller in the BMJ who points out that - even though only 51 volunteers started the trial - the statistics were analysed as if this number was 120, giving a completely misleading idea of the power of the study. If the statistics are done correctly the significance deteriorates from the claimed 5% to an entirely worthless 34%. As Miller says, “The only conclusion is that the trial is not able to prove anything”.

Brown, again in the BMJ, points out that it is unlikely the patients recruited into the trials even had perennial rhinitis in the first place, and Hadjicostas, a homeopathic researcher, comments (very politely), “I would like to say to the respectable members of the group who organized the research, that what they have done is not actually homeopathy...”, but says he feels this doesn’t matter since, what ever it was they were testing, it worked - and it’s results that count with homeopaths (at least when they are favourable to homeopathy anyway).

A similar tone is struck by Tim Vickers and Andrew Lancaster in their carefully balanced commentary at the end of the paper where they point out that, despite the authors' assertions (that this fourth paper added to cumulative proof that homeopathy was effective), the outcome measure - nasal peak inspiratory flow - looked at in the last paper wasn't significant in any of the other three so no such conclusion could be drawn. Quite the opposite actually, as Vickers and Lancaster say, “These data do not strengthen the conclusion that homoeopathy differs from placebo. In fact, the effect of including the current study in their meta-analysis with data from the three earlier trials is to weaken (though not overturn) this conclusion...”. Oops!

This conveniently flexible approach to outcome measures is another major criticism of the whole series. One poster in the JREF discussion comments that when the papers are studied carefully the authors seem to have done the trials, got the figures for several outcome measures and then chosen the best two to present in the conclusion. This tactic, known as the “Texas Sharpshooter”  fallacy, is akin to firing a load of bullets at the side of a barn, drawing a circle round the best group and then calling yourself a sharpshooter - it is almost impossible to get a bad result using this method (what that has got to do with Texas I don’t know - there are barns elsewhere in the world - but the analogy is a good one!).

So, apart from questionable statistics, flawed methodology, the possibility that the patients may not have been suffering from the condition supposedly under test and the opinion of a homeopath that treatment under trial wasn’t actually homeopathy and then anyway, even if we ignore all those points, this allegedly life-changing treatment still only barely manages to outperform a mere sugar tablet - just what is the problem with this series of trials?

Well, I'll tell you - it's a one off.

In the real world, if homeopathy was really as wonderful and effective as its proponents say, there would be dozens of such results. True, there is no such thing as a perfect trial, it's always possible to find one flaw or another but then again hayfever is an extremely common disorder and there is no shortage of potential experimental subjects looking for an effective, risk-free fix. Thus the body of evidence should have been mounting rapidly in the nearly-30 years since the original trial in this series.

But where is it? Why hasn't this work been repeated time and again by other independent authors, perhaps sponsored by those enormously rich homeopathic pharmacies, until the weight of evidence becomes so overwhelming proponents can legitimately dismiss any criticisms as unwarranted nit-picking?..

Homeopaths are fond of criticising opponents for not being open minded enough to believe that homeopathy might work. My question to them would be: are they open minded enough to admit that it doesn't? Because I’m afraid that is the only sensible answer to all of the questions above.

History of RVM How to use the site

The Best They Can Do The Usual Suspects How to look at evidence CAM journals and evidence Scepticism about homeopathy Arguments Acupuncture references