Having just co-authored a book on alternative medicine with the world's first professor of complementary medicine, Edzard Ernst, I have spent the last week fending off attacks from practitioners. Even though we endorse some alternative therapies, it seems that the alternative medicine community is unhappy with our conclusion that many others (such as reiki, homoeopathy, magnet therapy and crystal healing) offer nothing except a placebo effect. In other words, the supposed benefits of therapies such as homoeopathy are merely the result of wishful thinking.

In the case of homoeopathy, there have been more than 200 scientific trials and the overall conclusion is that homoeopathic remedies are nothing more than sugar pills. It is the expectation of recovery that boosts a patient's sense of wellbeing, and this is magnified by an encounter with an empathetic homoeopath in a relaxing environment. This explains why a bogus therapy can give the impression of being an effective medicine.

Despite all the evidence indicating that homoeopathic pills are placebos, practitioners continue to argue that this cannot possibly be the case. One of their most convincing claims, at least at a superficial level, is that many pet owners give their animals homoeopathic pills and they are convinced that they see remarkable improvements. Of course, the animals have no special expectation, so the placebo effect is irrelevant.So what is the explanation?

One possibility is that the pets make a real recovery soon after receiving a homoeopathic remedy, but the improvement is due to natural healing processes that would have taken place regardless of any intervention. The owner, who has put time, money and effort into providing a homoeopathic remedy, would rather give credit to homoeopathy than consider the natural recovery possibility. Hence, pet owners might be unreliable witnesses.

The only way to find out if homoeopathy really works on animals is to conduct a clinical trial, which means taking a large group of animals with a particular condition and giving them homoeopathic pills, while giving sugar pills to a parallel control group. The trial is conducted in a double-blind format, which means that neither the animals nor the vets know which creatures are receiving which treatment. This is revealed only when all the results have been gathered. This double-blinding reduces biases and leads to a more reliable result. The question being addressed in such trials is simple: does homoeopathy perform better than placebo sugar pills.

In 2003 the National Veterinary Institute in Sweden published a double-blind trial of the homoeopathic remedy Podophyllum, which was used as a cure for diarrhoea in calves. Twenty-four calves received homoeopathy and had diarrhoea for 3.1 days on average. Meanwhile, 20 received a placebo and they suffered for 2.9 days. In short, homoeopathy behaved just like a sugar pill.

More recently, a Cambridge University research group conducted a much larger trial. It involved 250 cows and compared homoeopathy against a placebo as a treatment for mastitis. The strength of such a trial is that there is an objective way of checking for improvement in inflammation of the udder, which is to count the number of white blood cells in a cow's milk. The conclusion, again, was that homoeopathy was no more effective than the placebo.

All in all, this means that homoeopaths who claim that their remedies work on animals are wrong. Fortunately the law allows only qualified vets to treat animals. Of course, homoeopaths are allowed to treat human beings, which means that sick animals have more protection under the law than sick humans when it comes to the ludicrous claims of alternative therapists.

Simon Singh is the co-author of Trick or Treatment? Alternative Medicine on Trial

(Bantam Press, £16.99)

Explore Health

• Expert Advice

• Health Features

• Mental Health

• Alternative Medicine

• Child Health