In this book an eminent physician explores how patients and caring doctors can help lessen suffering when illness occurs. Dr. Howard Spiro urges that physicians focus on their patients’ feelings of pain and anxiety as well as on physical symptoms. He also suggests that patients and their doctors be receptive to the emotional relief that may be obtained from nature and from hope.
Drawing on his previous highly praised work on the doctor-patient relationship and the problem of pain, Dr. Spiro tells how people can be helped by a combination of alternative medicine and mainstream medicinea treatment of mind, body, and spirit that energizes patients, strengthens their expectations, and starts them on the road to feeling better. In various forms of alternative medicine, from meditation to massage, from faith healing to folk medicine, from herbology to homeopathy, practitioners heed patients’ complaints and help them to help themselves. Dr. Spiro encourages physicians to talk and listen to their patients as much as they look and measure, to treat the whole patient and not just the disease, and to integrate a scientific approach to medicine with alternative approaches that may alleviate pain and suffering.
|Publisher:||Yale University Press|
|Edition description:||New Edition|
|Product dimensions:||5.00(w) x 8.50(h) x (d)|
About the Author
Howard Spiro, M.D., professor of medicine at Yale University School of Medicine and director of the Program for Humanities in Medicine there, established the gastrointestinal section at the Yale School of Medicine in 1955. He is the author of Clinical Gastroenterology, now in its fourth edition, and Doctors, Patients, and Placebos. He is coeditor of Empathy and the Practice of Medicine and Facing Death, both published by Yale University Press, and When Doctors Get Sick.
Read an Excerpt
The Placebo Drama
The placebo drama has four parts: a physician gives what he or she believes to be an inert material to a sick person, and the placebo effect follows--symptoms improve. Hidden behind the curtains, the placebo response, which accounts for the improvements, gets little applause. I believe the placebo response to be caused mainly by a change in the patient's perception. The physician, the placebo, and the patient together bring about the relief of symptoms that we call the placebo effect. Oral medicines--pills or potions--are the usual placebos, but operations, injections, and even diagnostic and therapeutic maneuvers also may be placebos. Many patients feel better after taking a placebo, but that does not mean that their disease, if they have one, has improved, for relief of symptoms is not the same as cure of disease.
The Oxford English Dictionary says that placebo was the name given to Vespers in the Office for the Dead because the first antiphonal response began with the Latin placebo ("I shall please"). Verse 9 of Psalm 116 in the King James version of the Bible goes, "I will walk before the Lord in the land of the living," but somehow an error turned "I will walk" into "I will please." So placebo gradually became a synonym for an inert or harmless medicine given to please the patient. In 1811 the term first came into use as "medicine prescribed more to please the patient than for its therapeutic effectiveness."
Henry Beecher, a Harvard professor who provided the foundation for most subsequent observations of the placebo and its effects, wrote of its common purpose "as a psychological instrument in the therapy of certain ailments arising out of mental illness, as a resource of the harassed doctor in dealing with a neurotic patient to determine the true effect of drugs apart from suggestion, in experimental work, as a device for eliminating bias not only on the part of the patients but also, when used as an unknown, of the observer, and, finally, as a tool of importance in the study of mechanisms of drug action" (Beecher 1955). Placebos are used today in two ways: in clinical trials as a substitute for treatment and in practice as therapy. These are quite separate purposes.
To evaluate the effect of any proposed new treatment, controlled clinical trials are needed. To be controlled, a clinical trial must have four features: (1) random assignment of subjects to therapeutic or control groups, (2) a double-blind structure, (3) placebo control to minimize bias, and (4) enough subjects to satisfy statistical requirements specified beforehand. Let me give an example.
Duodenal Ulcer Trials
To find out whether a new drug relieves pain or speeds healing of duodenal ulcers, a group of patients each of whom has dyspepsia and an ulcer crater seen at endoscopy receive a new drug. Because duodenal ulcers tend to heal on their own with almost any treatment or even no treatment, patients taking a new drug must be compared to others who receive no treatment. If patients know what they are ingesting, any improvement might be psychological, thanks to their faith in a new wonder drug or the extra attention lavished on them by the doctors. A control group therefore receives an inactive substitute identical in appearance to the active drug. The substitute is called a placebo.
To be fair to all, patients are randomly chosen by computer to receive either drug or placebo. They are informed about the selection process beforehand. Because the enthusiasm of the physician heightens the benefit of a drug as part of the placebo effect, and because chagrin at receiving no therapy or the same old standby might worsen indigestion in the control patients, such trials are carried out "double-blind"; physicians and patients alike are ignorant of who receives which treatment.
Placebos in such trials also act as a measure of the natural tendency of an ulcer to heal, especially since just being cared for may speed natural healing. That is an important consideration in placebo healing.
Controlled trials brought their own surprises twenty years ago: almost 60 percent of ulcers healed in placebo takers and only a few more, about 70 percent, in the specific-drug takers. An observer who concluded that the drug, cimetidine, was as effective as it proved to be, might have wondered about that 60 percent. What was healing by placebo telling physicians about the medical arts and the need for medications? Could the benefit of the new drug be almost incidental to the symbolic nature of the interaction between physician and patient, however constrained by double-blind conventions?
In the excitement about the new drug, however, most investigators ignored the benefit of the placebo, even though patients who got the placebo did almost as well as those who were given the active agent. At two weeks ulcers were documented to have healed in 56 percent of patients receiving the active drug, significantly more than the 37 percent of patients receiving the placebo, but at other periods up to six weeks cimetidine and placebo proved equally effective. Although a number of theories explained away these observations, there was little discussion about the effect of the controlled trials themselves or the specific helpfulness of placebos.
The dramatic placebo response of patients with many other disorders--the anginal chest pain of heart disease among them--makes it hard to sort out how much relief comes from the dedicated attention that patients enrolled in such studies receive (and could likewise receive outside a study from an ordinary office visit), their hope that they are receiving the active drug, and the ceremony of participating in a study.
Observers have wondered whether patients in such studies who fear getting a placebo would be skeptical about the promised benefit from even the active drugs. After all, they point out, an analgesic gave more relief in a controlled trial when the patients knew that it was being compared with another active drug than when they knew it was being tested against a placebo. That suggests that expectation brings relief, a crucial contribution to the placebo effect. Telling patients who are given a placebo that they are receiving an active drug might help them, too, but that would be unethical.
That tells us that controlled clinical trials with placebos are really conditioning experiments: those who receive the active drug get relief and therefore reinforcement every time they take it because of its genuine therapeutic effects. In contrast, those who take the placebo get no such reinforcement, because the treatment they receive is without intrinsic effect. Comparing one active drug to another also gives reinforcement, which may be why such trials usually generate enthusiastic comparisons of a new drug with the tried and true.
Some hold that just being selected for a trial helps. To evaluate that possibility, Irving Kirsch suggests that a group of patients be given neither the treatment nor the placebo, so that their improvement or lack thereof will tell whether expectation contributes to the placebo effect. After all, college students who expect to get high on beer quite often do so on just one glass if everyone else around them is drinking. Another way of evaluating the effect of expectancy is to give all three groups a standard established treatment but different expectations of just how much better they will feel.
Spontaneous remissions must account for some of the celebrated triumphs of placebos and other medical practices, complementary and mainstream alike. Editors of journals and newspapers, as well as patients and doctors, prefer accounts of miracles to those of failures, which explains many ephemeral "breakthroughs" in the treatment of cancer that fill the news media year after year.
How an observer affects a study has been called the Hawthorne effect: increased attention brings its own benefit to those being watched. Factory workers were more efficient when they thought they were being observed, just as doctors in an emergency room who were told that they were being monitored did better than when they were observed secretly. Thinking that they were being observed made doctors and factory workers alike work harder and better, just as visiting a doctor helps patients relax and begin to improve.
Ethics of Placebos in Trials
The ethics of controlled trials with placebos has had so much discussion that no general rule is ever likely to be set. Everyone agrees that humans should never be treated as a means to an end, but in medical practice, decisions are made on a case-by-case basis, in a casuistry sometimes decried.
It is hard to justify giving a child a placebo instead of a proven vaccine, but it may not be unethical to use a placebo on an adult who runs a very, very small risk without treatment. Informed consent is the rule; the patient-subject must give the final assent.
Open trials of new medications--that is, trials in which physicians and patients know who gets the placebo and who the active drug--are still debated despite the preceding considerations. Physicians rightly point out that they want to find whether a new drug has any potential value before they enlist patients for a more rigorous study. But if an open trial gives the impression that the drug is very effective, continuing a placebo control can be criticized.
Very few have found fault with placebos in clinical trials, unless there are already drugs so effective that not taking them will harm patients. Placebo trials may benefit others more than they benefit the 50 percent of patients in the study who turn out to be controls. Investigators want, and the Food and Drug Administration (FDA) demands, firm statistical data; a comparison with a placebo will give a dramatic contrast between the effect of a new drug and the natural course of a disease. To compare a new drug with a proven one requires far more patients. To some extent, this is a social question: How many resources should be used to prove the marginal superiority of a new drug over an old one? Indeed, when the omission of treatment will not harm patients, one may ask whether the disease needs any treatment at all.
What about those volunteers who have only a fifty-fifty chance of receiving effective therapy? That depends on the disease under study. For lymphoma, for example, where modern therapy has proven so lifesaving, any proposed treatment must be compared to current methods. For uncomplicated duodenal ulcers, which are generally benign, placebo therapy in a trial seems ethical to me, but not many other doctors agree these days. Doubts linger about whether placebo trials are ethical when known therapy has minimal side effects and those of the new agent are unknown.
Importance of Placebo Controls
Comparing one drug with another already in use may lead researchers to accept drugs as effective when they are not very active at all. Ulcers heal spontaneously, as we have seen--20 percent did in a London study, 70 percent in a Swiss study, and from 50 to 60 percent in a U.S. study. Such studies remain pertinent even if the reasons for the healing remain unclear--whether something in the patients, in their doctors, or in the general expectation of a cure. In one very old study, antacids relieved ulcer pain 79 percent of the time in one hospital and only 17 percent of the time in another. In France, the placebo-healing rates were different for different physicians but were fairly constant over time.
All those reports tell us that people react differently to placebos, whether from expectation or from the patient-physician relationship. The reports illustrate the uncertainties of comparing a new agent to one already known to be active. One group of patients may consist of more rapid responders than another--that is, have patients who are not quite so sick as in another group--so a drug equal to the standard may really be no better than a placebo.
The statistical difficulties are severe, too. It has been claimed by statisticians that seven hundred subjects are necessary to avoid false conclusions about the efficacy of any new drug. Now that peptic ulcers are so readily treated by acid-ablating measures or by antibiotics, enlisting that many patients to compare one new drug with a placebo is not feasible. Gastroenterologists are so convinced that antibiotics will eliminate ulcers forever that the question may never arise.
It is easier to agree when controlled trials with placebos are not in order: when (1) death or disability may result without treatment, (2) available therapy is very effective, or (3) the experimental drug is already proven effective and the trial is intended only to give exact numbers to the benefit for marketing reasons.
The ethics of clinical trials hang on the differences between patients and subjects. Many physicians feel like double agents when carrying out clinical studies on their patients, uncertain whether they are advancing their own interests or those of the patients. It may never be possible to procure from an acutely sick patient valid informed consent to any kind of therapeutic trial. The forty-year-old man in the cardiac intensive care unit finds it hard to refuse to participate in a study when his own doctor leans over the bedrail to ask for his cooperation.
Physician as Double Agent
More and more physicians were acting as double agents way before managed care. When a pharmaceutical firm pays a physician to run a controlled trial in a private office, the patients are almost never told about the financial arrangements, which are deemed not to be part of informed consent. It is no different for the assistant professor in a medical school who knows that promotion depends on publications; although he or she may not receive direct financial benefits from enrolling patients in studies, other benefits are forthcoming, such as secretarial help, travel, and other perquisites paid for from the commercial support to the laboratory or medical school. When federal agencies, like the National Institutes of Health (NIH), support similar studies, there are requirements to enroll a certain number of subjects in each study, and there are penalties for failing to achieve the appropriate "body count." In enrolling patients for studies, some academic physicians wonder whether they are considering their own welfare more than their patients'. The studies are important, and I have no quarrel with them; but any doubts could be resolved by disclosing all financial arrangements and any others to the patients enlisted? Whether those volunteer patients should also be paid needs far more discussion. Currently, for reasons that escape me, it is held unethical for doctors to pay volunteers to participate in studies. There are many other ways in which physicians can be double agents; the financial rewards for doing less rather than more are still large in managed care contracts.
The natural history of Crohn's disease, an inflammation of the intestines, underlines why controlled trials are so essential. The natural history of a disease can temper the doctor's enthusiasm for doing something and can highlight the benefits that come from self-help. "A patient sick enough to require treatment on an ambulatory basis can get better with no specific drug therapy," according to doctors at New York's Mt. Sinai Hospital. In one trial they reported improvement in 25-40 percent of the patients who received no treatment at all. More impressive, three-quarters of the patients in remission stayed well without specific maintenance therapy over the next year; two-thirds remained well for at least two years. Physicians used to know that many patients, even with Crohn's disease, get better on their own, without treatment. Patients in such studies have continuing physician support and enthusiasm, which may have a lot to do with their getting and staying well. That is what placebos and alternative medical approaches tell us. Unfortunately, physicians rarely recognize the healing powers of nature--or the passage of time.
Managed care administrators may cavil at too many visits, and physicians may worry that they are not really doing anything in follow-up visits, but experienced clinicians have long suspected that continuing contact with patients with Crohn's disease--and many other diseases, for that matter--prevents trouble. Patients must realize this too, for a number of people with Crohn's disease have asked for another appointment. That kind of medical contact may provide whatever placebos also provide, without medication.
Placebos as Therapy
Pharmacologically inactive, placebos have been called "a chemical device for psychotherapy, ... largely devoid of pharmacological properties." Shapiro's definition is more specific. He says a placebo is "any therapy (or that component of any therapy) that is deliberately used for its non-specific psychologic or psychophysiologic effect, or that is used for its presumed specific effect on a patient, symptom, or illness, but which, unknown to patient and therapist, is without specific activity for the condition being treated" (Shapiro 1964).
Others agree on the powerful psychological effects of "any therapeutic procedure ... which is objectively without specific activity for the condition treated." Howard Brody adds that "a placebo is used ... for its symbolic effect."
Therapeutic placebos relieve pain and the kinds of symptoms I described in Chapter 1. The responsible mechanisms may remain uncertain, but from the physician's standpoint, the medical intent is important even if one doctor's drug may be another's placebo. In evaluating the effect of placebos on disease, the physician's intent is important to definitions, if not to the patient, whose role in the placebo effect looms large.
Science and Intuition
Let me define science and intuition as I use those terms throughout this book. I will spend more time on intuition than science, which is so much part of our daily lives and thoughts that it has become almost a religion. Science is the process by which new knowledge is created; it gives objective knowledge that is quantitative and verifiable. The scientific method depends on observation, experimentation, and replication; therefore, science depends on sensory organs, especially the eyes, and on instruments that magnify and record what scientists can see.
Beyond experimentation and control, the scientific method requires verification by others. For Bertrand Russell, "Science is the attempt to discover by means of observation, and reasoning based upon it, first, particular facts about the world and then laws connecting facts with one another and (in fortunate cases) making it possible to predict future occurrences." He adds, "Science does not include art, or friendship, or various other valuable elements in life.... Science can tell us much about the means of realizing our desires, but it cannot say that one desire is preferable to another" (Russell 1935).
Intuition, in contrast, is what we know by immediate apprehension, at a glance, without conscious activity. I think of intuition as knowledge that is nonmeasurable and nonquantifiable. The Oxford English Dictionary defines intuition as "immediate knowledge.... The immediate apprehension of an object by the mind without the intervention of any reasoning process." The Random House Dictionary defines intuition as "direct perception of truth, fact, etc., independent of any reasoning process; immediate apprehension."
William Montague distinguishes intuition from science in his discussion of romantic love: "It would surely be a vain and preposterous undertaking to discover one's true sweetheart by accepting the authority of others, by using deductive reasoning and calculation, by cold-blooded empirical analysis of her perceivable qualities, or by considering the extent to which she might be a practical utility."
Henri Bergson depicts intuition and intelligence as pointing in opposite directions. I find his explication very helpful. Intelligence, he says, is the tool that science uses to deal with matter, with things and quantitative relations, whereas intuition is inward and immediate, providing a vision of reality: "The former toward inert matter, the latter toward life."
The theologian Martin Buber gives a rich interpretation of intuition: "We must develop in ourselves and in the next generation a gift which lives in man's inwardness as a Cinderella, one day to be a princess. Some call it intuition, but that is not a wholly unambiguous concept. I prefer the name `imagining the real,' for in its essential being this gift is not a looking at the other, but a bold swinging, demanding the most intensive stirring of one's being, into the life of the other."
Bertrand Russell derides intuition, comparing it to "mysticism." "When a man of science tells us the result of an experiment, he also tells us how the experiment was performed; others can repeat it, and if the result is not confirmed it is not accepted as true.... The mystic himself may be certain that he knows, and has no need of scientific tests but those who are asked to accept his testimony will subject it to the same kind of scientific tests as those applied to men who say they have been to the North Pole. I cannot admit any method of arriving at truth except that of science, but in the realm of the emotions I do not deny the value of the experiences which have given rise to religion" (Russell 1935).
For the scientist, intuition can lead to serendipity. The prepared mind is able to grasp a new opportunity; it has what the neurobiologist calls hard-wired circuits, ready to go with the proper calcium influx. For one retired physician, H. R. Jacobs, musing over the discoveries that had come early in his career, intuition was the "welcome stranger ..., essentially wild ..., invisible, inaudible, and imponderable. It is not subject to scientific testing because it is not amenable to scientific procedures; certainly it cannot be summoned at will."
In a sense, then, as Bergson suggests, science deals with matter, and intuition with mind, those two hoary philosophical poles. Someone has suggested that matter pushes, whereas mind pulls--a nice distinction.
The knowledge of science is quantifiable; the knowledge of intuition is not. This is the difference between intelligence and emotion, or physics and poetry. Poetic imagination sometimes may prove as illuminating to medical practice as science, but in judging the claims of alternative medicine, science is required. If two claims are made for cure of a disease, one by a faith healer and the other by a mainstream investigator, both must be assessed on a scientific basis. As Russell says, "What I do wish to maintain--and it is here that the scientific attitude becomes imperative--is that insight, untested and unsupported, is an insufficient guarantee of truth, in spite of the fact that much of the most important truth is first suggested by its means.... Instinct, intuition, or insight is what first leads to the beliefs which subsequent reason confirms or confutes" (Russell 1935). I will have more to say about intuition and the romantic stream in Chapter 12.
Physicians talk so much about "truth" and "facts" that it is worthwhile considering the Oxford English Dictionary definition of fact as "something that has really occurred or is actually the case ... a particular truth known by actual observation or authentic testimony, as opposed to what is merely inferred." Truth is defined, synonymously, as "conformity with fact."
In this regard, let me note the contrast between Yale's law school and its medical school. It often struck me during the five years that I had an office at the law school that Yale lawyers seemed to regard the law as created, not discovered, and as changing over time. The case method was much used in classes: the facts of a case were ascertained, and discussion was voluble. Then the instructor would suggest, "But supposing the man had been a woman or ...," changing first one aspect of the case, then another. The ensuing decisions varied depending on the changes in the details.
In contrast, in Yale's medical school, as in most others, students are taught that facts are immutable, that scratching away to find them will bring out the truth, and that there really is only one truth. We need both points of view as we contemplate placebos and alternative medicine and the benefits that they may bring.