hormone, which was, in fact, just a placebo. The injection site was painted with a disinfectant that left a prominent red stain, and the patients were injected twice a week for two weeks. By the third week, two of the three had broken their silence, and were talking quite sensibly.22
This study is, as far as I can tell, the only evidence that schizophrenia might be placebo-responsive, and it is not particularly strong evidence at that. For a start, there was no control group, the sample size was tiny – only three patients – and we do not know how long the improvement lasted. We cannot even be sure that the three patients were all really suffering from schizophrenia. Back in the 1950s, diagnostic standards were extremely variable, and the diagnosis of schizophrenia in particular was often used as a catch-all category to deal with any kind of patient whose prognosis was not good. Many people with learning disabilities, personality disorders or even simply unconventional behaviour were tagged as schizophrenics and confined to locked wards throughout Europe and America, where they eventually became so institutionalised that their behaviour mimicked the very diagnosis they had been given. They became mute and withdrawn, just as those suffering from the catatonic form of schizophrenia were supposed to. A vicious circle would establish itself, in which the condition of the patients and the pessimistic views of the nursing staff regarding their recovery became mutually reinforcing. It is not hard to see how, given a change in the views of the nursing staff, this circle could be broken. Perhaps this is what occurred with Lehmann’s patients. If so, it is not really a case of placebos curing schizophrenia.
In the absence of more conclusive evidence, then, it would be premature to pronounce any verdict on whether or not placebos can affect severe psychotic illnesses such as schizophrenia. For the time being, all we can say is that less serious mental conditions such as anxiety disorders and most types of depression do appear to be placebo-responsive. When it comes to schizophrenia, the jury is still out.
THE LIMITS OF THE PLACEBO RESPONSE
Tracing the limits of the placebo response is not easy. So far, we have seen that there is good evidence that it works for pain and some of the minor signs of injury, such as inflammation and trismus. There is also some evidence that placebos can alleviate depression and anxiety and cure ulcers. On the other hand, there is no real evidence that the placebo response can do anything to reverse cancer or cure schizophrenia. It would be interesting to know if there is any underlying logic to this. What is it that determines whether or not a medical condition responds to a placebo?
To answer this question, we must delve deeper into the underlying mechanisms of the placebo response. The placebo response is simply a rapid readjustment of the body’s own natural healing mechanisms to a surge of hope, and there are limits to what these mechanisms can do, even if fuelled with industrial-strength optimism. The placebo response is not magic; it works by natural mechanisms, and all such mechanisms have their limits. These mechanisms are the subject of the following chapter.
Chapter 3 THE ACUTE PHASE RESPONSE
It is all very well to know which conditions placebos can affect and which they can’t, but a good scientific theory requires more than just a list. Chemistry, for example, only became a true science when the great Russian chemist Dmitri Mendeleyev took the list of known elements and perceived a hidden pattern in the data. The periodic table, which Mendeleyev first formulated in 1869, made this pattern visible to all, and allowed scientists to uncover the fundamental order that underlay the mass of empirical results that they had accumulated in the previous centuries. To put the placebo response on an equally scientific footing, we must look beneath the signs and symptoms described in the previous chapter and ask whether there is some mechanism, or set of mechanisms, which explains why placebos can do the things they can, and why they can’t do the things they can’t.
At first sight, the list of conditions that placebos can alleviate may appear an odd bunch, lacking any common thread, as random as the elements must have appeared to those before Mendeleyev. Recent discoveries in immunology, however, have uncovered strong links between these conditions. It turns out that pain, swelling, stomach ulcers, depression and anxiety do, in fact, have something in common: they all involve the activation of a complex biological phenomenon known as the acute phase response. Perhaps this is the key to understanding the placebo effect.
THE ACUTE PHASE RESPONSE
Dealing with injury and infection is vital to survival. It is hardly surprising, then, that all animals possess mechanisms designed specifically to deal with wound healing and microbial defence. In mammals such as ourselves, these mechanisms are remarkably complex and, when they function correctly, produce an exquisitely choreographed suite of reactions which biologists are only now beginning to fully appreciate. The first stage in this process is known as the acute phase response, or, less technically, as inflammation.
For hundreds of years, Western medicine recognised the four signs of inflammation as tumor, rubor, calor and dolor – swelling, redness, heat and pain. In the last decades of the twentieth century, biologists discovered a few more. Besides these physical changes, there are also important psychological ones, including lethargy, apathy, loss of appetite and increased sensitivity to pain – a suite of symptoms that are collectively known as ‘sickness behaviour’.1 Taken together, the four classic signs of inflammation and the psychological symptoms of sickness behaviour constitute the complex set of processes referred to as the acute phase response.2
For a long time, doctors tended to regard the signs of inflammation and sickness behaviour as part of the disease process itself. The lethargy that commonly ensues after infection, for example, was thought to result from debilitation, as if the body had simply run out of energy. It is now known, however, that the various components of the acute phase response are not themselves pathological. On the contrary; they are actively produced by the body itself as part of the healing process. They may feel unpleasant, but they are actually good for you. In fact, feeling unpleasant is a vital part of their function.
PAIN
The value of feeling bad is nowhere better illustrated than in the case of pain. Pain, as everyone knows, is a great protector. The classic textbook illustration shows someone withdrawing their hand very rapidly from a hot stove they have just touched by accident. This acute pain is obviously beneficial, causing you to move away quickly from damaging objects. Even more important, however, is the second phase of pain that tends to follow the acute pain. Acute pain is sharp and stabbing, and ends when you are no longer in contact with the source of damage; the second type of pain is deep and spreading, and can last for minutes, hours, days, or even months. This kind of pain is not caused by pressure or heat from the outside world, but by chemicals released by the body itself. And, unlike acute pain, which produces rapid movement, the second type of pain causes you to keep the wounded area as still as possible, and encourages you to take extra care to shield the area from fresh injury while the process of repair is completed.
The capacity for pain, then, confers an advantage on those who have it. Exactly how advantageous this capacity is may be inferred from the sorry fate of those rare individuals who happen to be born without it. Almost everyone with this condition, which is known as congenital analgesia, is dead by the age of thirty. Without the capacity for acute pain, they do not withdraw from damaging objects unless their other senses inform them of the injury. One girl with the disorder, for example, knelt on a hot radiator while she gazed out of a window, blissfully unaware of the burning flesh. Her knees were scarred for the rest of her life, which was not very long, since she died at the age of twenty-two.3
It is not the lack of acute pain that kills people with congenital analgesia, however, but the lack of the second kind of pain. The lack of the deeper, longer-lasting second phase