CONCLUSION T O PA R T 1: THE PLA CEBO EFFECT Put simply, the fundamental puzzle about medicine from the ﬁfth century bc until the end of the nineteenth century is that doctors found patients who were prepared to pay for treatment that was at best ineﬀectual, and usually deleterious. Throughout this period sur- gery, particularly abdominal surgery, was commonly fatal, while the most common therapies were bloodletting, purging, and emetics, all of which weakened patients. Advances in knowledge, such as the discovery of the circulation of the blood, had no pay-oﬀ in terms of advances in therapy, so that we might say that all the progress was in human biology, none of it in medicine. At the end of the seventeenth century, Charles Perrault, in his Parallèle des anciens et des modernes, felt quite safe in having the spokesperson for the ancients dismiss all modern anatomical discoveries as irrelevant to treatment.
Looking back to the beginning of the century, Arthur Hertzler, a Kansas doc- tor, wrote in 1938, ‘I can scarcely think of a single disease that the doctors actually cured during those early years . . . The possible exceptions were malaria and the itch. Doctors knew how to relieve suﬀering, set bones, sew up cuts and open boils on small boys.’ In 1905 Joseph Matthews, a past-president of the American Medical Association, felt that the only drugs physicians really needed were laxatives and emetics. But the fact that there was no progress –– far too little to have any systematic impact on life expectancy –– and the fact that medical intervention did more harm than good, does not mean that doctors did not cure patients. Modern studies of the placebo eﬀect show that it is a mistake to think that there are some therapies that are eﬀective and others which though ineﬀective work on those who respond to the placebo eﬀect. Even eﬀective medicine works partly by mobilizing the body’s own resources, by invoking the placebo eﬀect: one estimate is that a third of the good done by modern medicine is attributable to the placebo eﬀect. When patients believe that a therapy will work, their belief is capable of rendering it surprisingly eﬃcacious; when doctors believe a therapy will work their conﬁdence is consistently trans- ferred to the patient. There are all sorts of studies that show this in practice. Thus if a new and better drug comes out, the drug it replaces begins to perform consistently less well in tests, merely because doctors have lost conﬁdence in it. If doctors administer eﬃcacious drugs believing them to be placebos, then their eﬀectiveness is less than if they administer them believing them to be eﬃcacious.
If you change the size or the colour of a pill, or the number of times a day it is administered, you alter its eﬀectiveness. Patients who faithfully follow their doctors’ instructions do better than those who do not, even if the pills they are being instructed to take are inert. It is important though to stress that if Hippocratic medicine beneﬁted its patients by mobilizing the placebo eﬀect, Hippocratic therapies were not in themselves placebos. Placebos are inert sub- stances, and new drugs are regularly tested against placebos in blind clinical trials –– a drug has to outperform a placebo before it is regarded as having any therapeutic eﬀect. But Hippocratic remedies were far from inert. Bloodletting, purging, and emetics acted power- fully and, in so far as they acted on the body, they were bad for patients. In so far as they acted on the mind they may have been good for patients, but we can be conﬁdent that if one tested Hippocratic remedies against placebos the placebos would outperform the Hippocratic remedies: doing worse than a placebo is, if you like, a technical deﬁnition of what I am calling ‘bad medicine’ or ‘doing harm’. By this deﬁnition, which I think is the appropriate one, you are doing harm even if your patient is more likely to recover as a result of receiving your treatment than if he had received no treat- ment at all, providing your treatment is less beneﬁcial than a placebo. The doctor Foucault tells us about, who abandoned Hippocratic remedies and gave all his patients quinine, was giving them a better conclusion to part i: the placebo effect 69 therapy, not because quinine is eﬀective, but because he was coming closer to giving a true placebo. Homeopathy, founded by Samuel Hahnemann, who published his Handbook of Rational Healing in 1810, is regarded by conventional modern doctors as working by mobilizing the placebo eﬀect, and homeopathic remedies are held to be inert substances.
What hom- eopathy can do for you is thus a good indication of what a placebo can do for you, and, on the deﬁnition I am proposing homeopathy does neither good nor harm, though it is perfectly reasonable that it should be available (as it increasingly is) on the National Health Service, since it performs much better than no treatment at all. Homeopathy, we can be sure, would have outperformed Hippocratic medicine in a trial. It follows, then, that for the ﬁrst hundred years or so homeopathy was superior to conventional medicine; it is only for the last hundred years that conventional medicine has had a strong claim to be superior to homeopathy. We can be conﬁdent then that medicine has always been better for patients than no treatment at all, but until the late nineteenth century the beneﬁt of treatment usually derived solely from the fact that doctors and patients believed it would be beneﬁcial, and consequently it was. We can also be clear that the type of beneﬁts that medicine was capable of oﬀering, until the last century, and leaving aside some simple surgical procedures and a very few other treatments, was eﬀectively restricted to what the body is capable of doing for itself. Thus if a patient takes a placebo believing it to be a pain-killer they are likely to experience a reduction in pain, and this reduction is not just in the mind: the body produces endorphins, which reduce the pain. In this way the placebo can mimic the working of opiates. But the body is incapable of producing a substance comparable to aspirin (introduced in 1899), so that even if you take a placebo believing it to be aspirin, the body will never successfully mimic the action of aspirin; your pain relief will still come from the production of endorphins. Before 1865, as after, doctors were able to marshal all the resources of the placebo eﬀect, and it is a safe general rule, to which there were, as Hertzler acknowledged, very few and very limited exceptions (setting bones, reducing dislocations, operating for bladder stones and cataracts, and, in later periods, taking opium for pain relief, quinine for malaria, digitalis for dropsy, mercury for syphilis, orange and lemon juice for scurvy), that this is all that medicine could do. For more than two thousand years medicine eﬀectively stood still, despite all the progress in human biology, and a doctor in ancient Rome would have done you just about as much good as a doctor in early nineteenth-century London, Paris, or New York. But if modern medicine is eﬀective and Hippocratic medicine was not, it follows that the very idea that there is continuity between the two is profoundly misleading.
The same institutions may educate doctors in the twenty-ﬁrst century as in the thirteenth (you can still get medical degrees at Bologna, Paris, Montpellier); many of the same words may be used to describe diseases; but modern medicine is no more a development of ancient medicine than modern astronomy is a development of medieval astrology. The two are fundamentally diﬀerent. At the very beginning of the twentieth century, on the other hand, the medical care that could be oﬀered by doctors such as Hertzler and Matthews was still essentially Hippocratic. Bleeding had been largely abandoned (though it continues to be used even at present for two fairly rare conditions, hereditary haemochromatosis, a disease of excessive iron storage in the body, and polycythemia, a complication of severe lung disease), for it made no sense for those diseases that were now thought of as infections; and humoral theory had given way to modern physiology; but laxatives and emetics were still the doctor’s basic remedies. Twenty years earlier, and the standard therapies were even closer to those of Hippocrates: in 1878 Émile Bertin, in his article on bleeding in the Dictionnaire encyclopédique des sciences médicales, is still recommending bloodletting for a wide range of conditions, and supporting his argument by appeals to Galen and Thomas Sydenham, ‘the English Hippocrates’. This was still bad medicine. Bertin’s patients would have been better advised to go to a homeopath.