Trick or Treatment (19 page)

Indeed, James Randi has continued to offer his $1 million to anyone who can independently reproduce the effects claimed by Benveniste. BBC television took up the challenge as part of its
Horizon
science documentary series, gathering together a team of scientists to oversee the project. They examined the effect of a homeopathically diluted
histamine
on cells, and compared this with the effect of pure water. Histamine is associated with allergic responses in cells, but would it still cause cells to react if it had been diluted to the extent that it was no longer present? Professor Martin Bland of St George’s Hospital Medical School announced the final result: ‘There’s absolutely no evidence at all to say that there is any difference between the solution that started off as pure water and the solution that started off with the histamine.’ As anecdotal evidence to reinforce the point, Randi mentioned the following story during the programme: ‘I also consumed sixty-four times the prescribed dosage of homeopathic sleeping pills and didn’t even feel drowsy. I did this before a meeting of the US Congress – if that doesn’t put you to sleep, nothing will.’

While biologists were trying and failing to find evidence for homeopathy acting at a fundamental cellular level, physicists tried to examine homeopathy at a basic molecular level. It was clear that ultra-dilute homeopathic solutions contained only water and no molecules of the active ingredient, but some physicists wondered if the water molecules somehow had altered their arrangement in order to retain a memory of the earlier ingredient.

Over the last two decades, physicists have published the results of dozens of experiments examining the molecular structure of normal water versus homeopathically prepared water. They have used powerful and arcane techniques such as nuclear magnetic resonance (NMR), Raman spectroscopy and light absorption to look for the slightest evidence that water has a memory of what it once contained. Unfortunately, a review of these studies published in
The Journal of Alternative and Complementary Medicine
in 2003 showed that these experiments were generally of poor quality and prone to errors.

For example, one NMR experiment claimed to detect a difference between the molecules in ordinary water and those in a homeopathic remedy, but in the end this was attributed to a problem with the equipment. The NMR apparatus is supplied with test tubes made of soda glass, which is not a very stable form of glass. Hence, when the homeopathic solution was shaken during its preparation, glass molecules were leached into the solution. Not surprisingly, this homeopathic solution responded differently to the pure water in terms of its NMR profile, which initially gave the misleading impression that the homeopathic solution was demonstrating a water-memory effect. Sure enough, when another research team repeated the experiment with borosilicate glass test tubes, which are much more stable than soda glass, the NMR instrument could no longer detect any difference between water and homeopathic remedies. Yet again, experiments have so far failed to find anything surprising about the behaviour of molecules in homeopathic solutions.

In summary, homeopaths have been disappointed that physicists probing water molecules have found nothing special about homeopathic remedies. Similarly, biologists looking at single cells have not made any great breakthroughs in finding convincing evidence that might support homeopathy.

All of this, however, matters very little in terms of the main homeopathy debate, because what happens at the molecular or cellular level is of much less interest than what happens to patients. Forget biology or physics, because homeopathy is all about medicine. The fundamental question is straightforward: does homeopathy heal patients?

Homeopaths, of course, have always been confident that their remedies cured a range of symptoms, but in order to persuade doctors and everyone else that homeopathy was truly effective, they needed concrete evidence from scientific trials. We have explained in previous chapters that the most conclusive type of clinical trial is the randomized, placebo-controlled, double-blind trial; if such trials could generate results that supported Hahnemann’s ideas, then this would force the medical establishment to embrace homeopathy. Alternatively, if these studies failed to show that ultra-dilute solutions offered any benefit, then this would mean that homeopathy was nothing more than quackery. As the twenty-first century approached, rigorous trials were about to be conducted on a massive scale. The results would eventually settle the debate over homeopathy once and for all.

Homeopathy on trial

For many homeopaths, the lack of definitive scientific evidence to support their remedies was not a matter of concern, because they could cite numerous examples that seemed to demonstrate the effectiveness of their interventions. For instance,
The Complete Idiot’s Guide to Homeopathy
by David W. Sollars includes a story told by a mother who treated her son, Kailin. The boy had burned his arm at a barbecue, but luckily the people hosting the occasion had just purchased a homeopathic home kit:

This case seems impressive, but it was dissected and undermined by Jay W. Shelton, author of
Homeopathy
. He identifies four questions that challenge the significance of this and similar cases. First, this seems like a classic first-degree burn, the least serious type, which causes damage only to the surface of the skin; so should we really be so surprised about the lack of blisters? Second, why should homeopathy get any credit, when the recovery process might have been entirely due to the body’s natural healing ability? Third, is it possible that the ice-filled glass played the most important role in helping the child? Finally, if homeopathy did indeed help the patient, then could its influence have been entirely due to the placebo effect? In the previous chapter we saw the power of placebo, which is so great that it can make a useless therapy appear like a truly valuable one, as long as the patient has confidence in the therapy.

When scientists dismiss such human cases as partly or wholly due to the placebo effect, homeopaths often cite cases of animal healing, because they believe that animals are immune to the placebo effect. It is true that many pet owners and farmers feel that homeopathy helps their animals, and it is also true that these creatures are oblivious to what a pill is supposed to do, but the value of these anecdotes collapses under closer scrutiny.

For instance, the animal is unaware of which treatment it is receiving or how it is supposed to respond, yet the fact remains that the person monitoring the animal is fully aware. In other words, the animal is effectively blinded to what is happening, but the person reporting the events is unblinded and is therefore unreliable. For instance, an anxious pet owner who has faith in homeopathy might focus on any sign of improvement based on expectation and hope, while ignoring symptoms that have deteriorated. Even if the animal has definitely improved beyond the placebo effect, then this could be due to a variety of factors other than the homeopathic pill, such as the extra care and attention being provided by a caring and concerned owner.

In short, the medical establishment will not accept anecdotal evidence – based on either human or animal patients – as reliable enough to support homeopathy or any other treatment. No amount of anecdote can stand in place of firm evidence, or, as scientists like to say, ‘The plural of anecdote is not data.’

Medical scientists place an emphasis on data because the best way to analyse the impact of any therapy is to look at the results from rigorous scientific investigations, particularly clinical trials. By way of a quick recap, you will recall that Chapter 1 revealed the uncanny ability of the randomized clinical trial to show which therapies work and which do not. Chapter 2 then built on that foundation to show how this technique could be used to test the claims of acupuncturists. So what happens when homeopathy is submitted to the same scientific scrutiny?

In theory it should be much easier to test homeopathy than acupuncture, because it is much more obvious how to take into account the placebo effect. A homeopathic trial would require the random assignment of patients into two groups, namely a group treated homeopathically and a placebo control group. The patients would not be told to which group they had been assigned. Both groups would receive an empathic encounter with a homoeopath, who would also be blinded, inasmuch as he or she would not know which patients belonged to which group. Researchers would then create two batches of pills that were identical, except that one batch would have been treated with a drop of homeopathic solution and the other batch would remain plain. The treatment group would receive the homeopathic pill and the control group would receive the plain pill. Patients in both groups should experience some improvement, simply due to the placebo effect. The critical question is this: does the treatment group on average show significant improvements over and above the control group? If the answer is ‘yes’, then this would clearly indicate that homeopathy is genuinely effective. If, however, the answer is ‘no’ and each group shows a similar response, then homeopathy would be exposed as having nothing more than a placebo effect.

Before looking at the trials conducted with humans, it is interesting to note that there have been some randomized placebo-controlled trials of homeopathy’s impact on animals. The overall conclusion of the major studies is that homeopathy offers no benefit to animals. For example, in 2003 the National Veterinary Institute in Sweden conducted a double-blind trial of the homeopathic remedy
Podophyllum
as a cure for diarrhoea in calves, and it found no evidence for the efficacy of homeopathy. More recently, a Cambridge University research group conducted a double-blind trial to compare homeopathy against a placebo as a treatment for mastitis for 250 cows. An objective way of checking for any improvement in inflammation of the udder is to count the number of white blood cells in the cow’s milk, and the conclusion was that homeopathy was no more effective than the placebo.

When scientists looked at the evidence in terms of human patients, the picture was more complicated. The good news was that by the mid 1990s there had been well over 100 published trials seeking to decide the therapeutic value of homeopathy. The bad news was that this mountain of research consisted largely of poorly conducted trials, often with inadequate randomization, or with no proper control group, or with insufficient numbers of patients. None of these trials was able to give a definitive answer to whether or not homeopathy benefited patients beyond the placebo effect.

With nothing to rely on except unconvincing anecdotes and inconclusive trials, the arguments for and against homeopathy were deadlocked. Then, in 1997, an international research team took a dramatic step towards settling the homeopathy debate. They were led by Klaus Linde, a senior figure at the Munich-based Centre for Complementary Medicine Research. He and his colleagues decided to examine the considerable body of research into homeopathy in order to develop an over-arching conclusion that took into consideration each and every trial. This is known as a
meta-analysis
, which means an analysis of various analyses. In other words, each individual trial into homeopathy concluded with an analysis of its own data, and Linde was proposing to pool all these separate analyses in order to generate a new, more reliable, overall result. Meta-analysis can be considered as a particular type of systematic review, a concept that was introduced in the previous chapter. Like a systematic review, a meta-analysis attempts to draw an overall conclusion from several separate trials, except that a meta-analysis tends to involve a more mathematical approach.

Although the term meta-analysis might be unfamiliar to many readers, it is a concept that crops up in a range of familiar situations where it is important to make sense of lots of data. In the run-up to a general election, for instance, several newspapers might publish opinion polls with conflicting results. In this situation it would be sensible to combine all the data from all the polls, which ought to lead to a more reliable conclusion than any single poll, because the meta-poll (i.e. poll of polls) reflects the complete data from a much larger group of voters.

The power of meta-analysis becomes obvious if we examine some hypothetical sets of data concerning astrology. If your astrological sign determined your character, then an astrologer should be able to identify a person’s star sign after an interview. Imagine that a series of five experiments is conducted around the world by rival research groups. In each case, the same astrologer is simply asked to identify correctly a person’s star sign based on a five-minute conversation. The experiments range in size from 20 to 290 participants, but the protocol is the same in each case. Chance alone would give rise to a success rate of one correct identification (or hit) in twelve, so the astrologer would have to do significantly better than this to give credence to the notion of astrology. The five experiments lead to the following success rates: