Extraordinary claims require extraordinary evidence. Homeopathy makes perhaps the most extraordinary claims of any branch of alternative medicine, yet the extraordinary evidence required for such claims has not been provided. In fact, homeopaths have trouble providing ordinary evidence that their treatments work. What they are quite good at is providing excuses…
Claim – Homeopathy works
Evidence – Despite being utterly implausible, trials have been conducted (to the dismay of those who feel that testing such obvious nonsense is futile). Large, well-designed trials have found no benefit above placebo. Systematic reviews of well-designed trials have been conducted and there is even a systematic review of systematic reviews. The bottom line? Homeopathy doesn’t work. Homeopathic products are inert.
Excuse #1 – You can’t test homeopathy in a randomised controlled trial, as it is individualised.
The truth – You can test individualised homeopathy using an RCT. Not only is it possible to do this, it has actually been done. In 1991, Kleijnen et al found 14 trials of individualised homeopathy. In 1998, Linde and Melchart conducted a review of randomized controlled trials of individualized homeopathy and found that “when the analysis was restricted to the methodologically best trials no significant effect was seen” (the authors concluded that “The results of the available randomized trials suggest that individualized homeopathy has an effect over placebo. The evidence, however, is not convincing…“). [See also Ben Goldacre‘s comments on individualised homeopathy in the Guardian.]
Excuse #2 – Randomised, placebo-controlled clinical trials of individualised homeopathy fail because placebo effects are substantially higher than in conventional medicine.
The truth – In 13 matched sets the placebo effect in the homeopathic trials was larger than the average placebo effect of the conventional trials, in 12 matched sets it was lower (P=0.39). Additionally, no subgroup analysis yielded any significant difference. Placebo effects in RCTs of individualised homeopathy were not larger than in trials of conventional medicine.
Claim – Diluting remedies makes them more powerful. Even when they’re so dilute that there is not a single molecule of the ‘active’ ingredient left.
Evidence – This claim contradicts everything from common sense to the laws of physics and chemistry. If dilute remedies were so powerful, then we’d be easily able to observe their effects in properly conducted trials. We can’t. In fact, homeopathy doesn’t work. Highly diluted homeopathic products are not more powerful, they are inert.
Excuse – Never mind the laws of physics. There doesn’t need to be any active ingredient present. Homeopathy works by the memory of water.
The truth – Researchers have found that “liquid water essentially loses the memory of persistent correlations in its structure within 50 fs”. There is some comment here on the ‘memory of water’ and homeopathy:
…the ‘explanations’ on offer seem either to consider that water physics can be reinvented from scratch by replacing decades of careful research with wishful thinking, or they call on impurities to perform the kind of miraculous feats of biomolecular mimicry and replication that chemists have been striving to achieve for many years.
Claim – Effects of homeopathy cannot be explained by the placebo effect.
Evidence – Homeopathy can be explained by the placebo effect. “…there was no homeopathic remedy that was demonstrated to yield clinical effects that are convincingly different from placebo”.
Excuse – Homeopathy works in children and animals, therefore it cannot be placebo.
The truth – There are several possible reasons for the perceived improvements in animals treated with homeopathy, all of which are more likely reasons for the perceived improvement than homeopathy. For example, spontaneous improvement, fluctuation of symptoms, regression to the mean, additional treatment… and the placebo effect.
I’m not sure why some people believe that placebo effects cannot be produced in children or animals. Not only have those making this argument failed to provide me with evidence that the placebo effect is irrelevant to children and animals, they have also failed to explain the ‘logic’ behind their thinking.
For placebo in animals, there may be an expectation on the part of the owner or vet that the treatment will be beneficial and there may be a conditioning effect on the animal being treated. If Clever Hans could respond to subtle visual clues in order to ‘perform arithmetic’ and pigeons could be trained to recognise a target and peck at it, then is it really so implausible that a cow might respond to the body language of a farmer or vet who was enthusiastically treating the animal with homeopathy? If adult humans can be fooled into thinking that a treatment has made them better, is it really so implausible that the same adult humans can be fooled into thinking a treatment has led to improvements in the symptoms of an animal they are treating?
As to whether children can benefit from a placebo effect… Children with drug-resistant partial epilepsy receiving placebo in double-blind RCTs demonstrated significantly greater 50% responder rate than adults, probably reflecting increased placebo and regression to the mean effects.