Come in Homeopathy! Your Time is Up!

" ‘Alternative' medicine is usually defended by a 'skeptical' argument, that we should keep our minds open." Petr Skrabanek in his article "Demarcation of the absurd"1 looked for guidelines on quite how open we should leave our minds and for how long. As he put it "Anything is possible. 'You have to keep your mind open'—until your brains drop out." He argues that we should be prepared to express unbelief because we can always change our minds, but by being gullible or keeping the mind too wide open we "lose reason from the very beginning."

Francis Crick expressed what I think is a similar view, though on a different subject, that of religious belief: "Humanists on the other hand, consider these beliefs to be false. If pressed they may say that they are agnostic about them, and that these matters cannot at the moment be finally decided one way or the other. In less cautious moods they will admit that they consider these...ideas to be nonsense, and too silly for any scientifically educated person to believe in at this time."

I suspect that homeopathy's time for similar treatment has come; and to it one might add colour acupuncture therapy, iridology and diagnostic aspects of electroacupuncture-according-to-Voll, but in this essay I will concentrate on homeopathy which has had since about 1790 to prove itself. In about that year, Samuel Hahnemann, justifiably disillusioned with the medicine of his time, hit upon the principle of "similia similibus" or "like through like". A patient's symptom or symptom complex should be treated with a medicament which-itself produces those symptoms. "Sympathetic medicine" had in fact been mentioned in the Hippocratic collection, but had been lost by the time of Galen. Hahnemann believed there was an initial effect of the medicament followed by an after effect whereby vitality is restored and sets up a defence against the "alien" agency. If the initial effect were too strong, the healing effect could be ruined and this led him to dilute the medicine to the point at which there was no initial effect to counter the healing effect.

He would serially dilute a drop of plant juice or other substance one in ten up to thirty times, vigorously shaking the mixture between each dilution, believing not only that the potency increased with dilution, but that the shaking or "succussion" caused the remedy's "ghostlike mysterious powers to be released and to increase". Insoluble substances were dealt with by grinding them with sugar (trituration). Hahnemann spent many years experimenting upon himself and members of his family, taking a dose of a substance and then noting the symptoms that ensured during the following thirty or forty days. This process was known as "proving". Between about 1805 and 1810, he incorporated the results of his experiments in a book: Organon der rationellen Heilkunde now known simply as the "Organon".

Compared to the purging, enemas, blood-letting and other foul remedies-of-his contemporary physicians, Hahnemann's remedies were considerably more pleasant to take and, it has to be said, harmless, so that homeopathy rapidly acquired a following, especially perhaps in France where giving enemas was something of an art form, and to where he moved to spend the rest of his life from 1835 to 1843.

Thus the principles of homeopathy include taking a careful history of the patient's symptoms, choosing an appropriate substance or substances which has been proved to cause those symptoms, diluting it up to one part in 10%, succussing it at each dilution to increase its potency and giving it to the patient for as long as necessary.

The reader might suppose that this is all very interesting history but in fact it appears to be how homeopathy is practiced nowadays except that most of the proving has been done. After taking a careful history of symptoms in much more detail than would a science-based physician the homeopathic physician then examines the patient in much the same way as his colleagues and if a science-based remedy such as surgery, radiotherapy, or modern drug is appropriate he prescribes it. He may feel that it would be appropriate to add a homeopathic remedy to the other therapy or that a homeopathic remedy alone is what is needed. In this casé he consults an edition of the Organon and prescribes a homeopathic remedy. He tries to match the remedy exactly to the symptoms, as one of the core beliefs in homeopathy appears to be that the illness is the symptoms rather than the illness being the cause of the symptoms.

A quotation from a recent clinical trial of homeopathy in the treatment of hay fever by Reilly e¢ a/.3 will illustrate how the remedy is prepared. "A homeopathic pharmacist impregnated 125 mg lactose tablets...with 0.1 ml of mixed grass pollens 30c... (to prepare a 30c potency one drop of grass pollen extract in 99 drops of 90% alcohol-in-water solution is succussed. One drop of the product is then placed in 99 drops of fresh dilutent and this is succussed. This process is done thirty times, to produce a 30c potency, which, were it expressed as a dilution, would be one part in 10°°)". Those who have forgotten might like to be reminded that Avogrado's number, the number of molecules in a gram molecule of a substance is 6 x 10".

At this stage it is easy to declare that homeopathy cannot work, that the facts speak for themselves and so on, since the probability of a 30c "potency" containing even one molecule of the medicament is negligible. Modern homeopaths accept that this is so but claim that their methods work through an as yet unexplained mechanism.

Is there any hard evidence that it does work? There have been a few controlled trials of homeopathy against placebo published in reputable medical journals but all the ones in English I have been able to track down3,4,5 have been criticised on a variety of grounds, for example, having inadequate controls due to inappropriate selection of patients6 or inappropriate choice of placebo7. One re-designed trial8 which attempted to answer some of the criticisms showed a small improvement in symptoms in favour of homeopathy, but the redesigned trial was not itself free from flaws.9 Several critics have complained of "lack of scientific validity" or in effect said that homeopathy cannot work and so does not work. I have to confess that I would feel very uneasy if an impeccably designed trial were to show an unequivocal and reproducible effect.

The study of Reilly et al.3 comes close to being impeccable except that the inactive substance ("control") administered to half the patients should have been identical in every way to the allegedly active medicine except for the initial presence of the pollen extract. It was not, as the control was not diluted and succussed in the same way as the "active" substance. In the normal course of events this would be considered to be a trifling criticism, but we are being asked to accept something extraordinary and the standard of proof required is naturally expected to be very high indeed.

Reilly showed that he did not accept that the control was not adequate and in doing so tied himself into a logical knot by implying that if the control had becn treated in the same way, it might have become "potentised" and the issue confused.10

In 1984 Day reported in a veterinary journal on the control of stillbirths in pigs using homeopathy.11 It is a little hard to understand why it was published at all as the author himself writes "Accepting the limitations imposed by failure to set up a double blind trial and the fact that the litter size was smaller in the test group, the results suggest that the treatment was efficacious." It is precisely because of these limitations that the study failed to show anything of scientific worth. The same author published again in 1986, this time in a homeopathic journal12, on homeopathic treatment of bovine mastitis. He admitted statistical weaknesses "...but underlying trends appear too powerful to ignore them totally". In the late 1930's medical papers began to use tests of statistical significance although it was not for another twenty-five years that it became commonplace and then almost obligatory to do so because of the recognition of observer bias and how it can lead to trends being perceived where none exist. Science based medicine has its problem areas too in that some specialities "...have all the qualities of pseudoscientific dogma and unproven efficacy apparent in alternative medicine: the appeal to personal experience and anecdote is remarkably similar in the two cases."13 This does not of course validate alternative medicine.

Apparently a quarter of doctors in France practice homeopathy to some extent14 and 15 percent of remedies supplied are homeopathic.15 In the face of the increasing popularity of homeopathy, the French Ministry of Social Affairs has set up a group to carry out controlled clinical trials. It recently reported that it had found no effect of homeopathy on the time taken for the intestines to start working again after abdominal surgery.16 Two of the authors of this report, Poitevin and Benveniste, also shared in the authorship of a now celebrated paper published in Nature in June of this year.17 This paper from Professor Benveniste's laboratory purported to show that when cells bearing IgE antibodies on their surface are exposed to anti-IgE antibodies, changes can be demonstrated in the cells even when the anti-IgE antibodies are diluted beyond the point at which any molecules can remain. The "solution" needed vigorous shaking for the effect to be observed. The amount of change waxed and waned as the solution got progressively more dilute.

Unusually, Nature printed an editorial disclaimer, "When to believe the unbelievable,16" which pointed out that it would be premature and probably mistaken for homeopathic medicine to see the paper as giving aid and comfort, because the results "'strike at the roots of two centuries of observation and rationalisation of physical phenomena." It would be "prudent to ask more carefully than usual whether the observation may be incorrect." In a letter to Nature shortly after, Reilly had the misfortune, for a homeopathic physician, to write "If we now rise to this challenge, we have nothing to lose. If we prove the observations wrong, we will have exposed homeopathy as one of medical sciences greatest misadventures.19" On the next page, Fierz pointed out that some of the data "seemed very unlikely to be derived from flawless experimental results...20" as the experimental errors normally expected to be present were absent from the published results.

Nature had made publication of the original article conditional on allowing a team to visit the laboratory where the experiments had been carried out, to observe procedures and to examine laboratory records. Maddox, the editor of Nature, Randi, a professional magician and Stewart, a biochemist, concluded that the claims made by Benveniste and his colleagues "'are based chiefly on an extensive series of experiments which are statistically ill-controlled, from which no substantial effort has been made to exclude systematic error, including observer bias, and whose interpretation has been clouded by the exclusion of measurements in conflict with the claim that anti-IgE at 'high dilution' will degranulate basophils.21" When an experiment did not "work" the data were recorded but not included in the analyses prepared for publication; and this explained Fierz's observation that the experimental results as published were surprisingly free from error, an observation also made by Gaylarde.22 When observer bias was carefully excluded, the results could not be reproduced.

By 4 August 1988, only the smile was left. Metzger and Dreskin briefly reported a failure to replicate Davenas et al. claim using a closely related system23 and several chemists had suggested how some of the observations might have been obtained, thus accommodating the results within the existing scientific paradigms.24,25,26 Benveniste had found the visit of the team from Nature very upsetting, feeling that "a tornado of intense and constant suspicion, fear and psychological and intellectual pressure unfit for scientific work" swept the lab.2? He felt that he and his colleagues were under suspicion of having cheated and appeared unable to accept Randi's explanation"! that sensational claims could be sustained only by data of exceptional quality. Much had been made in the popular press of the discovery that the salaries of two of the authors of the study were paid by the manufacturers of homeopathic remedies, but Benveniste pointed out quite fairly that the same is true in many other areas of research.

If homeopathy is more effective than placebo how might it work? Reilly suggested “...some workers have theorised that the succussion induces electrochemical patterning of the diluent which can then replicate at every stage of the potentising process. One physicist presented evidence...suggesting that succussion produces energy storage in the bonds of the diluent in the infra-red spectrum which 'downloads' in contact with the water in living systems. Perhaps this information then spreads like a 'liquid crystal' through the body water, modifying receptor sites or enzyme action...nuclear magnetic resonances work.,.claimed to distinguish between simple dilutions and succussed solutions.3" Scientists in general have remained silent on this suggestion, dismissed it as nonsense28 or have expressed profound skepticism18, Reilly restated and refined his hypothesis in Nature this year! when commenting on the work of Davenas et al.17 While chemists, physicists and the like found plenty to say about the flaws in the work of Davenas et al., none of the many thousands of professional scientists who make up the readership of Nature have so far been moved to comment about Reilly's hypothesis, possibly finding it too incomprehensible for words.

Does homeopathy really matter? I am not all that happy about over-the-counter sales of homeopathic remedies as I think the public are then being gulled rather expensively. When practiced by a homeopathic physician we may forget for the moment Skrabanek's unkind words that "At present, the difference between a doctor and a quack lies not in the nature of their practice but in the possession of a medical diploma1", for conditions more appropriately treated by science-based medicine will not be overlooked and if all that homeopathy does is act like a powerful placebo it may well make the patient feel better. However, it does nothing for the image of primary care physicians as thoughtful, scientific healers, the best of whom have always practiced "holistic" medicine. It may show them as adherents of the new science of non-science (perhaps we should drop the hyphen) which demands it be investigated according to different criteria from hard science29; and if it does so it does harm to the whole profession. Lloyd Franklin is quite right that we may have modest pride in today's medical profession30. I have never seen a new case of polio, diphtheria, or syphilis. The elimination of smallpox remains a little-known triumph. It is unlikely that I will see another case of tuberculosis in the twenty or so years of medical practice remaining to me. These tremendous advances owe nothing to homeopathy. If it has any influence on the course of diseases it must be a small one as it is proving very difficult to demonstrate scientifically.

The controversies about homeopathy and other areas of non-science medicine prompt one to wonder to what extent the study of scientific subjects leads to any understanding of the philosophy of science. Few medical graduates will do research and even fewer will do research of high quality that advances human understanding significantly, but is it too much to hope that doctors do not employ the same fallacy-ridden reasoning as their patients and non-science colleagues? The medical curriculum has little space for extras but, once qualified, doctors often have little time for reflection, so that if science-based medicine is to continue to thrive, space must surely be found for explaining why people are able to hold bizarre beliefs and how we make sound, rational advances in medicine.

REFERENCES

  1. Skrabanek, P. Demarcation of the absurd. Lancet, 1986;

1:960-61.

  1. Crick, F. Letter. Varsity, 1966; November 12.
  2. Reilly, D.T. et al. Is homeopathy a placebo response.

Lancet, 1986; 2:881-86.

  1. Gibson, R.G. et al: Salicylates and homeopathy in rheu-_

matoid arthritis: preliminary observations. Br. J. Clin. Pharmacol. 1978; 6:391-95.

  1. Shipley, M. et al. Controlled trial of homeopathic treatment of arthritis. Lancet, 1983; 1:97-98.
  2. Huston, G. Salicylates and homeopathy. Br. J. Clin. Pharmacol. 1979; 7:529-30.
  3. Khan, M.F. Letter. Lancet 1986; 2:1107.
  4. Gibson, R.G. et al. Homeopathic therapy in rheumatoid arthritis: evaluation by a double blind clinical trial. Br. J. Clin. Pharmacol. 1980; 9:453-59.
  5. Couzigou et al. Letter. Lancet, 1983; 2:482.
  6. Reilly, D.T. et al. Letter. Lancet, 1986; 2:1272.
  7. Day, C.E.J. Control of stillbirths in pigs using homeopathy. Veterinary record. 1984; 114:216.
  8. Day, C. Clinical trials in bovine mastitis. British Homeopathic J. 1986; 1:11-14.
  9. Lancet 1983; 2:773. Alternative Medicine is no Alternative (Editorial).
  10. Nature 1988; 334:367. When to publish pseudo-science (Editorial).
  11. Plasterk, R.H.A. Letter. Nature, 1988; 334:285.
  12. Mayaux, M.J. et al. Controlled trial of homeopathy in post-operative ileus. Lancet, 1988; 1:
  13. Davenas, E. et al. Human basophil degranulation triggered by very dilute antiserum against IgE. Nature, 1988; 333:816-18. ;
  14. Nature, 1988; 333:787. When to believe the unbelievable (Editorial).
  15. Reilly, D.T. Letter. Nature, 1988; 334:285.
  16. Fierz, W. Letter. Nature, 1988; 334:286.
  17. Maddox et al. "High dilution" experiments a delusion. Nature, 1988; 334:287-290.
  18. Gaylarde, P.M. Letter. Nature, 1988; 334:375.
  19. Metzger, H. and Dreslin, S.C. Letter. Nature, 1988;

334:375.

  1. Suslick, K.S. Letter. Nature, 1988; 334:375-6.
  2. Glick, J.L. Letter. Nature, 1988; 334:376.
  3. Escribano, M.J. Letter. Nature, 1988; 334:376.
  4. Benveniste, J. Dr Jaques Benveniste replies: Nature, 1988; 334:291.
  5. Skrabanek, P. Letter. Lancet, 1986; 2:1107.
  6. Tizard, M.H. A comment on the 2,4,5-T Report. J.Gen.Practice, 1986; 3:4.
  7. Franklin, L.M. New Zealand Skeptic, 1986; No.4:3-4.