The demise of homoeopathy?

The two main principles of homoeopathy, namely the law of similars and the notion that highly potentised (diluted) remedies can be effective even though they are unlikely to contain a single molecule of the original substance, have remained controversial throughout their 200-year history. Opponents quite simply insist that where there is no molecule there can be no effect, and that all clinical effects of homoeopathy must therefore be due to placebo. A meta-analysis of all randomised or placebo-controlled trials, however, concluded that the clinical effects of homeopathy are not entirely due to placebo.1
This meta-analysis has frequently been criticised, eg, for pooling together data relating to all types of indications and remedies.2,3
Such an evaluation, one could argue, is akin to a meta-analysis of all conventional drugs ever used for any condition. The result would probably be “positive” (ie, it would show that drugs are significantly better than placebos) but, at the same time, utterly meaningless. Which drug? Which condition?

It is therefore relevant to assess defined indications and remedies and see what evidence for or against homoeopathy emerges. The remedy that has been submitted to more controlled clinical trials than any other homoeopathic medicine is Arnica Montana. A recent systematic review of all studies of homoeopathic Arnica did not provide conclusive evidence that it has clinical effects which differ from those of placebo.4
This overall result was recently confirmed by independent researchers employing a similar methodological approach.5
The condition most frequently employed for testing homoeopathic remedies is delayed onset muscle soreness (DOMS). A systematic review of all of these trials did not produce convincing evidence that homoeopathic remedies are superior to placebo in treating this condition.6

But in their daily routine, homoeopaths rarely treat DOMS; they often aim to alleviate ailments like asthma or headaches. Thus it would be perhaps more valuable to assess the effectiveness of homoeopathy for these clinically relevant conditions. Two systematic reviews found no good evidence to suggest that homoeopathic remedies are efficacious for either of these illnesses.7,8
To make things worse, the above-mentioned meta-analysis comprising all randomised or placebo-controlled trials was re-evaluated. This time, much stricter entry criteria were used: for instance, trials had to be randomised and placebo-controlled and only highly diluted homoeopathic remedies were considered. The resulting meta-analytic effect size was almost precisely zero.9
Even the original authors of the above-named meta-analysis1
seem to have somewhat changed their minds. In a recent re-evaluation of their data they found a negative correlation between the methodological quality and the effect size.10
This suggests that trials of top quality will also tend to produce negative results.

Is this then the end of homoeopathy? Has it now been shown beyond reasonable doubt that homoeopathic remedies are mere placebos? Many would disagree, and certainly the general public does. In 1997, worldwide sales of homoeopathic products were estimated to be in excess of $1bn.11
A recent survey shows that in the US the popularity of homoeopathy is increasing.12
Homoeopaths argue that the integration of homoeopathy into routine health care would save large sums of money.13
In Britain, 92 per cent of GPs seem to be willing to refer patients for homoeopathic treatment.14

More importantly, new support for homoeopathy also comes from recent basic research. In an international collaborative study, four independent laboratories have found evidence that, in an in vitro test system involving human basophil degranulation, highly dilute homoeopathic remedies do elicit different responses to dilutant alone.15
This seems to justify Benveniste who had originally shown similar effects but was subsequently ostracised from the scientific community.16
Moreover, it was repeatedly demonstrated in various animal models that ultra-high homoeopathic dilutions seem to have significant biological effects.17,18
And, finally, it has been pointed out that the two above-mentioned axioms are not so implausible after all. The author points out that recent conventional research indicates that concentrations as low as 10?22 mol/L can be biologically active.19

The demise of homoeopathy does not therefore seem to be in sight. Both opponents and proponents probably feel that the clinical efficacy question has been answered definitively, albeit with opposite conclusions. Undoubtedly it would help homoeopathy if a mechanism of action was discovered and homoeopathy would no longer be a prime example of implausibility. However, the definitive answer, in my view, has to come from a series of rigorous trials. Both camps would have to agree to the protocols beforehand. The trials themselves ought to be carried out by fully independent investigators who have no axe to grind. I would argue that, in the interest of science, truth and our patients, this would be an exercise well worth doing.

Professor Ernst is head of the department of complementary medicine at Exeter University



1. Linde K, Clausius N, Ramirez G, Melchart D, Eitel F, Hedges LV, Jonas W. Are the clinical effects of homoeopathy placebo effects? A meta-analysis of placebo-controlled trials. Lancet 1997;350:834-43.

2. Vandenbrouk JP. Homoeopathy trials, going nowhere. Ibid 1997;350:824.

3. Langman MJS. Homoeopathy trials, reasons for good ones but are they warranted. Ibid1997;350:825.

4. Ernst E, Pittler MH. The efficacy of homoeopathic Arnica. A systematic review of placebo-controlled clinical trials. Arch Surg 1998;133:1187-90.

5. L?dtke R, Wilkens J. Klinische Wirksamkeitsstudien zu Arnica in hom?opathischen Zubereitungen. In: Albrecht H, Fr?hwald M (eds). Jahrbuch Band 5 (1998) Karl und Veronica Carstens-Stiftung. KVC Verlag Essen 1999:97-112.

6. Ernst E, Barnes J. Are homoeopathic remedies effective for delayed-onset muscle soreness? A systematic review of placebo-controlled trials Perfusion 1998;11:4-8.

7. Linde K, Jobst KA. Homoeopathy for chronic asthma. Cochrane Library 1998;1:1-7.

8. Ernst E. Homoeopathic prophylaxis of headaches and migraine? A systematic review. J Pain Sympt Managem. In press.

9. Ernst E. Are highly dilute homoeopathic remedies placebos? Perfusion 1998;1:291-2.

10. Linde K, Scholz M, Ramirez G, Clausius N, Melchart D, Jonas WB. Impact of study quality on outcome in placebo-controlled trials of homoeopathy. J Clin Epidemiol 1999;52:631-6.

11. Information Access Company. France leads world in homoeopathy. Market letter, May, 1996.

12. Eisenberg D, David RB, Ettner SL, Appel S, Wilkey S, van Rompay M, Kessler RC. Trends in alternative medicine use in the United States, 1990-1997. JAMA 1998;280: 1569-75.

13. Jacobs J, Chapman EH, Crothers D. Patient characteristics and practice patterns of physicians using homoeopathy. Arch Fam Med 1998; 7: 537-40.

14. Wyllie M, Hannaford P. Attitudes to complementary therapies and referral for homoeopathic treatment. Br Homoeopath J 1998;87:13-16.

15. Belon P, Cumps J, Ennis M, Mannaioni PF, Sainte-Laudy J, Roberfroid M, Wiegant FAC. Inhibition of human basophil degranulation by successive histamine dilutions: Results of a European multi-centre trial. Inflammation Research. 1999;48:S17-S18.

16. Fisher P. The end of the Benveniste affair? Br Homoeopath J 1999;88:186-7.

17. Datta S, Mallick P, Khuda Bukhsh AR. Efficacy of a potentized homoeopathic drug (Arsenicum Album-30) in reducing genotoxic effects produced by arsenic trioxide in mice: Comparative studies of pre-, post- and combined pre- and post-oral administration and comparative efficacy of two microdoses. Comp Ther Med 1999;7:62-75.

18. Jonas W, Lin Y, Williams A, Tortella F, Tuma R. Treatment of experimental stroke with low dose glutamate and homoeopathic Arnica Montana. Perfusion 1999. Inpress.

19. Eskinazi D. Homoeopathy re-revisited. Is homeopathy compatible with biomedical observations? Arch Intern Med 1999;159:1981-7.

Last updated
The Pharmaceutical Journal, PJ, January 2000;():DOI:10.1211/PJ.2000.20000060

You may also be interested in