Sham Surgery Returns as a Research Tool (April 25, 1999)
In the summer of 1994, a surgeon named J. Bruce Moseley found himself engaged in an elaborate form of make-believe. Moseley had 10 patients scheduled for an operation intended to relieve the arthritis pain in their knees. The patients were men -- most of them middle-aged, all former military guys -- and they weren't ready to consign themselves to the rocking chair yet. So they had decided to take a risk and volunteer for a study that must have sounded, when Moseley first told them about it, rather peculiar. All 10 would be wheeled into an operating room at the Houston Veterans Affairs Medical Center, draped, examined and anesthetized. All 10 would be dispatched to the recovery room and sent home from the hospital by the next morning equipped with crutches and a painkiller. But there the similarities ended. For while two of the men would undergo the standard arthroscopic surgery for their condition -- the scraping and rinsing of the knee joint -- and three would have the rinsing alone, five would have no recognized surgical procedure at all. Their surgery would be a placebo, an exercise in just pretend.
Moseley would stab the placebo patients' knees three times with a scalpel -- to make it feel and look real, there had to be incisions and later, scars -- but that was it. And he couldn't break character. If he knew in advance which kind of surgery he was to perform, he might somehow give it away, so it wasn't until he entered the gleaming O.R., scrubbed and in his greens, that he opened an envelope telling him whether he was doing a real procedure or a fake one that time. Only the anesthetist and the nurse assisting him were in on the secret.
In some ways, Bruce Moseley was an unlikely actor for such a role. As a surgeon and a team physician for the Houston Rockets, he wasn't the kind of doctor given to woolly introspection about the Mind-Body Problem. He was simply skeptical about the specific benefit of arthroscopic surgery for arthritis of the knee and wanted to test its efficacy. Then Moseley started talking with a doctor named Nelda Wray who had been put in charge of health care research at the Houston V.A., and she asked him a startling question: How did he know that whatever benefit came from this surgery wasn't a product of the placebo effect -- that is, that those who improved did so not because the operation actually healed the knee joint but because they expected it would?
"I said: 'It can't be,'" Moseley recalls. "'This is surgery we're talking about.' And she said: 'You're all wrong. The bigger and more dramatic the patient perceives the intervention to be, the bigger the placebo effect. Big pills have more than small pills, injections have more than pills and surgery has the most of all.'"
Moseley was intrigued, but before he could begin his study, it had to be approved by the institutional review board at Baylor and by the human-studies committee at the V.A. "My first reaction was what anybody's would be," says Baruch Brody, an ethicist who consulted with the committees. "I said: 'You're going to do what? Cut people up, pretending to do surgery when you're not?'" But Brody, like the rest of the reviewers, eventually came around to the idea. Arthroscopic knee surgery typically requires general anesthesia, which carries risks; the placebo procedure required only a sedative and was cheaper. If patients responded just as favorably to a placebo, why continue with arthroscopy?
The placebo worked. Six months after surgery, the 10 patients still didn't know whether they had been faked out or not. But all of them reported much less pain. None were unhappy with the outcome of the operation. This was a pilot study -- far too small to offer any definitive conclusions about the efficacy of arthroscopic surgery for osteoarthritis. But it was suggestive enough to set Moseley and Wray on an expanded version of the same research design -- this time with 180 patients -- for which they expect to have the results next October.
Sylvester Colligan, a 76-year-old World War II veteran from Beaumont, Tex., was assigned to the placebo group in Moseley and Wray's expanded study. Colligan's right knee had been giving him trouble for five years when a doctor at his local clinic told him he probably had arthritis and suggested he talk to Moseley. "I was very impressed with him, especially when I heard he was the team doctor with the Rockets," says Colligan, a courteous, soft-spoken retiree who worked for the local Coca-Cola bottling plant, first as an assistant to the driver of a delivery truck and eventually as an account manager. "So, sure, I went ahead and signed up for this new thing he was doing." Colligan doesn't sound all that at ease with the term placebo, but he does know his surgery consisted of only shallow incisions. More important, he knows that he has no pain in his knee now and that he can mow his yard again and walk wherever he wants. "The surgery was two years ago and the knee never has bothered me since," he says. "It's just like my other knee now. I give a whole lot of credit to Dr. Moseley. Whenever I see him on the TV during a basketball game, I call the wife in and say, 'Hey, there's the doctor that fixed my knee!'"
That operating-room fakery should exert a therapeutic effect on a patient is a remarkable notion. And yet it may be that the symbolic armature of surgery -- the shedding of blood, the cultural prestige of surgeons, even the scars that call to mind a dramatic act of healing -- is itself a powerful force in recovery. It may be that "placebo surgery" isn't such an oxymoron after all.
Forty years ago, a young Seattle cardiologist named Leonard Cobb conducted a unique trial of a procedure then commonly used for angina, in which doctors made small incisions in the chest and tied knots in two arteries to try to increase blood flow to the heart. It was a popular technique -- 90 percent of patients reported that it helped -- but when Cobb compared it with placebo surgery in which he made incisions but did not tie off the arteries, the sham operations proved just as successful. The procedure, known as internal mammary ligation, was soon abandoned. Over the next decades, the whole idea of placebo surgery fell out of favor with ethicists and patient advocates. If placebo surgery didn't violate the Hippocratic edict "First, do no harm," what did?
Now, though, that taboo is fading, and Moseley and Wray are not the only ones arguing that new surgical techniques, like new medicines, ought to be tested with placebo controls. Doctors investigating a new technique for treating Parkinson's disease, for example, incorporated a placebo control group -- patients who had holes drilled in their skulls to simulate the real procedure, in which fetal cells are implanted in the brain -- into a widely publicized study released last April. The study showed a significant placebo effect, though the motor ability of the patients who actually got the fetal implants improved even more. "This is just the beginning," Dr. C. Warren Olanow, chairman of neurology at Mount Sinai School of Medicine, told Time magazine. "Tomorrow, if you have a new procedure, you will have to do a double-blind placebo trial."
Some doctors argue, with reason, that the new enthusiasm for placebo surgery is driven by hospital bean counters and insurers who want hard evidence that an expensive procedure works before they'll pay for it. And yet, with all due respect for medical economics, that can't be the only explanation. If placebo surgery makes sense, it makes sense because of the growing body of evidence for the strength of the placebo effect in general. Maybe pretend operations are just the most sensational proof of a diffuse phenomenon that has never quite been explained, but that, as the historian of science Anne Harrington has written, continues to "haunt our house of biomedical objectivity."
What if today's H.M.O. doctors were not so circumspect about offering magic, hope and comfort? What if they resorted more often to the old-fashioned squeezing of the hand, the slightly unrealistic imparting of hope that is now dismissed as authoritarian and paternalistic?
The truth is that the placebo effect is huge -- anywhere between 35 and 75 percent of patients benefit from taking a dummy pill in studies of new drugs -- so huge, in fact, that it should probably be put to conscious use in clinical practice, even if we do not entirely understand how it works. For centuries, Western medicine consisted of almost nothing but the placebo effect. The patient who got better after a bleeding -- or a dose of fox lung, wood lice, tartar emetic or any of the other charming staples of the 19th-century pharmacopoeia -- got better either in spite of them or because of their symbolic value. Such patients believed in the cure and in the authority of the bewigged gentlemen administering it, and the belief gave them hope and the hope helped make them well. There were exceptions -- remedies, like quinine for malaria, the vaccine for smallpox and morphine for pain relief, that actually worked. But generally speaking, if all the drugs of the day "could be sunk to the bottom of the sea," as Oliver Wendell Holmes observed in 1860, "it would be all the better for mankind -- and all the worse for the fishes."
In the postwar era, with the triumph of the biomedical model and the proliferation of drugs and techniques that did successfully extend life, many doctors would just as soon have forgotten about the placebo effect altogether. They now had proven remedies and methods: antibiotics for infection; dazzling surgical techniques like heart and liver transplantation; sophisticated diagnostic tools like magnetic resonance imaging. In fact, medical science has improved so much and so fast in the last 40 years that it is easy, perhaps, for doctors to neglect the part of medicine that is not science at all. The ready and lavish display of sympathy, the laying on of hands, the projection of a slightly mystical authority -- these are now more often the province of alternative medical practitioners, who have no compunction about manipulating them. In a recent edition of a British medical journal, a general practitioner wondered whether he ought to give an 86-year-old patient a pharmacologically useless "tonic" she had requested in fond remembrance of the ones her old doctor had given her. "I don't use tonics routinely anymore," he wrote, "but perhaps I should: they are pharmacologically pretty harmless and cheap to boot. Homeopathy, herbs and talkative middle-aged ladies who squeeze your feet have replaced them, but because I do not offer these alternative medicines, my patients go elsewhere for their magic, hope and comfort."
But what if doctors firmly in the fold of evidence-based medicine were not so circumspect about offering magic, hope and comfort? What if they did resort more often to tonics or their equivalent -- to the old-fashioned squeezing of the hand and the slightly unrealistic imparting of hope that is now frowned upon as a bit too authoritarian and paternalistic? What if they started thinking of placebos as a way of bridging the gap between the magnificent but sometimes cold efficiency of modern American medicine and the unproven but evidently comforting remedies prescribed by homeopaths and herbalists? After all, more and more people seem to be drawn to alternative medical therapies -- half of all Americans say they have tried them -- despite a distinct lack of empirical proof for their effectiveness. Yet people may actually harm themselves if they down the latest untested herbal panacea, while if they take an inert substance (or get a cheering speech) given by a conventional physician, they will do no worse and they stand a good statistical chance of doing better. There is something to be said, surely, for a little benign deception.
Though a few doctors are starting to argue for a more conscious use of the placebo effect, even for the deliberate prescribing of dummy pills, this is not an easy case to make. Placebos have long slunk around with a reputation as nuisances -- necessary, but annoying. Most of us know them as the pharmacologically inert drugs given to some of the subjects in so-called double-blind studies of new drugs, in which neither the subjects nor the researcher know who is getting what. If as many subjects improve by taking the placebo as do by taking the active drug, then the active drug is thought to be a bust. It happens all the time, even to highly touted new medications. Last summer, the share price of a British biotech company called Peptide Therapeutics dropped 33 percent after it revealed that its new allergy vaccine was only as effective as a placebo. During the trials on food-allergy patients, a company spokesman had reported delightedly, 75 percent had improved to the point where they could tolerate foods they'd never been able to before. But when the control group data came in, so, awkwardly enough, had 75 percent of the subjects taking inert tablets. Last year, the pharmaceutical company Merck announced that it was halting development of MK-869, a new antidepressant it had been promoting for months as a blockbuster drug on the scale of Prozac. Alas, the dummy pills worked just as well. And in a recent study on VEGF, a genetically engineered heart drug announced with much fanfare by its manufacturer, Genentech, the placebo actually performed better . Two months after their treatments, patients who had gotten low doses of VEGF could walk 26 seconds longer on a treadmill, those who had gotten high doses could walk 32 seconds longer and those who had gotten a placebo could walk -- go figure -- 42 seconds longer.
Photograph by Craig Cutler for The New York Times
Of course, patients are often far from pleased to hear how well their placebos have worked. "Once we did a PMS study where we treated people for a month with placebo and then told people who'd responded what they'd been on," says Karen Weihs, a clinical psychiatrist at George Washington University. "And as it turns out, it's a very difficult thing to confront somebody with. Some people feel insulted, or silly. You're telling them it's all in their mind. We try to frame it positively -- your symptoms aren't so severe that you need medication; your mind has other ways of making you feel better. But that doesn't always help."
As if all this weren't stigma enough, there's the history of the placebo to taint it, too. The very word, Latin for "I shall please," carries a faintly unwholesome connotation. It's the first utterance you hear in the Catholic Vespers for the Dead, and in the Middle Ages, it referred, unflatteringly, to professional mourners who sang at funeral masses -- toadies, sycophants. A medical dictionary of 1811 defined a placebo as something "given more to please than to benefit the patient." Placebos were the sugar pills, the drops of colored water, the "tonics" dispensed to placate querulous or malingering patients. In the 19th and well into the 20th century doctors may have handed them out "by the bushels," as the eminent American physician Richard Cabot once admitted he did, but they also regarded them with some embarrassment. Placebos smacked of deception and quackery, and what's worse, upended the logic of Western medicine. "We all trust our sensation as a reflection of objective reality, and yet the placebo changes the sensation without affecting the objective reality," lamented Patrick D. Wall, a pain specialist. (This was in an essay with the pithy title "The Placebo Effect: An Unpopular Topic.")
Meanwhile, alternative-medicine types, who might be expected to think a little more loosely about such matters -- and to acknowledge a role for psychological states in physical healing -- are generally too wary of seeing their methods dismissed as "mere" placebos to take much positive interest in the placebo effect itself. Dr. Howard Brody, a professor of medicine at Michigan State University who is the author of a coming book on placebos, tells a story of two Korean acupuncturists who met at a conference of alternative practitioners and chatted amicably until they discovered that each practiced a slightly different form of acupuncture. They then viciously accused one another of purveying a placebo.
And yet the inconvenient evidence keeps trickling in that if placebos are lies, they can also be, in the words of one commentator on the phenomenon, "lies that heal." In an influential article first published in 1955, the Harvard researcher Henry Beecher concluded that between 30 and 40 percent of any treated group would respond to a placebo. But studies since then have shown placebos working for certain conditions -- pain, depression, some heart ailments, gastric ulcers and other stomach complaints -- in closer to 50 or 60 percent of subjects, sometimes more. Indeed, it's not unheard of for placebo effects to exceed those attributed to the active drug.
And strangely, the placebo effect is not limited to the subjective sensations of patients; some studies show actual physiological change as a result of sham treatments. Doctors in one study successfully eliminated warts by painting them with a brightly colored, inert dye and promising patients the warts would be gone when the color wore off. In a study of asthmatics, researchers found that they could produce dilation of the airways by simply telling people they were inhaling a bronchiodilator, even when they weren't. Patients suffering pain after wisdom-tooth extraction got just as much relief from a fake application of ultrasound as from a real one, so long as both patient and therapist thought the machine was on. Fifty-two percent of the colitis patients treated with placebo in 11 different trials reported feeling better -- and 50 percent of the inflamed intestines actually looked better when assessed with a sigmoidoscope.
Photograph by Craig Cutler
for The New York Times
How much of this sort of improvement can be written off to spontaneous remission of an illness or regression to the mean (some people would have gotten better anyway; many illnesses wax and wane) and how much of it lasts (studies are often only 8 or 12 weeks long and placebo effects may flag sooner than "real" ones) are still open questions. Because we treat so many illnesses aggressively, we know less about their natural history -- what would happen if we did nothing. Very few studies compare a placebo group with a group receiving no treatment at all, since that would tell us more about the placebo than the active drug, and it is the active drug most researchers care about.
"Some of what we call the placebo effect can surely be explained as the natural course of an illness," says Dr. Walter A. Brown, a professor of psychiatry at Brown University who has written extensively on the phenomenon. But not all of it, Brown believes. "There is certainly data that suggest that just being in the healing situation accomplishes something. Depressed patients who are merely put on a waiting list for treatment do not do as well as those given placebos. And -- this is very telling, I think -- when placebos are given for pain management, the course of pain relief follows what you would get with an active drug. The peak relief comes about an hour after it's administered, as it does with the real drug, and so on. If placebo analgesia was the equivalent of giving nothing, you'd expect a more random pattern."
With these kinds of observations to bolster his case, Brown has lately been making a highly unorthodox suggestion. If placebos accomplish so much in a research setting, why not let doctors deliberately prescribe them? Of course, the ethics of informed consent would require that a physician tell the patient what he was giving him. Brown, who thinks such a strategy would be appropriate for conditions like mild to moderate depression, pain, asthma and hypertension, in which a placebo has shown to be especially potent and in which distress plays an aggravating role, imagines the doctor saying something like this: "Mrs. Jones, the type of depression you have has been treated in the past with either antidepressant medicine or psychotherapy, one of the talking therapies. These two treatments are still widely used and are options for you. There is a third kind of treatment, less expensive for you and less likely to cause side effects, which also helps many people with your condition. This treatment involves taking one of these pills twice a day and coming to our office every two weeks to let us know how you're doing. These pills do not contain any drug. We don't know exactly how they work; they may trigger or stimulate the body's own healing processes. We do know that your chances of improving with this treatment are quite good. If after six weeks of this treatment you're not feeling better we can try one of the other treatments."
Brown's proposal, which he has been airing for the past few years in venues ranging from the journal Hospital Practice to USA Today, has touched off a spirited debate, and it's easy to see why. In the eyes of some critics, there is something shady about the crafty bedside pitch he has scripted, one that seems intended to mislead as much as inform. Donald Klein, a psychiatrist at Columbia University, worries that deliberately prescribing placebo antidepressants would "play into the hands of those who depict psychiatrists as artful, exploitative manipulators who take advantage of the patient's gullibility."
There's a reason Brown's script is so subtle. Plenty of evidence exists that placebos can work in a clinical setting so long as at least one party believes they are the genuine article. Therapies that have since been wholly discredited -- from mammary ligation for angina to milk-quaffing for ulcers -- produced precisely the results they were supposed to until they were shown to have no scientific basis for doing so. But can a placebo ever make you better if you know it's a placebo? Doesn't its agency lie only in the reasonable expectation -- the 50-50 chance, at least -- that you're getting the real thing? Does a placebo cease to be a placebo when you call its bluff?
Maybe the best way to consider whether Brown's idea would work in practice is to apply it to depression. That was the ailment for which he first suggested deliberate placebo treatment, back in 1994, in an article in the journal Neuropsychopharmacology. Since then, evidence that depression is an especially placebo-sensitive condition has only been mounting. The splashiest comes from Irving Kirsch, a psychologist at the University of Connecticut, who contends that the multibillion-dollar success of Prozac and its brethren may be attributed almost entirely to the placebo effect. In a study published this past June, Kirsch and his co-author, Guy Sapirstein of the Westwood Lodge Hospital in Needham, Mass., analyzed 19 clinical trials of antidepressants and concluded that the expectation of improvement, not adjustments in brain chemistry, accounted for 75 percent of the drugs' effectiveness. Kirsch says a study he is working on now, based on the clinical trials that won F.D.A. approval for the drugs Prozac, Zoloft, Effexor, Paxil and Serzone, bears out the earlier one. "The critical factor," says Kirsch, "is our beliefs about what's going to happen to us. You don't have to rely on drugs to see profound transformation."
Kirsch thinks his research justifies a round of somewhat unorthodox studies in which half of the people getting the active antidepressant would be led to believe they were getting the placebo. "It would involve some temporary deception and then a debriefing," he says. "But I think it's acceptable because it would be the way to show a true drug effect. Do people who are taking the drug but think they are taking the placebo do as well as people getting the placebo? The companies don't know and they don't seem to want to know."
Plenty of informed people -- and not just shills for the pharmaceutical industry or dedicated Listeners to Prozac -- think Kirsch goes too far. It's as if "he set out to prove that all antidepressant medication is a sham," says Dr. Steven Hyman, director of the National Institute of Mental Health, "and the evidence is not on his side there." Hyman points out that for ethical reasons, very few of the patients selected for outpatient antidepressant trials are severely depressed. No honorable researcher who thinks a patient might be suicidal, for instance, is going to sign him up for a study where he may well be getting only a dummy pill. That leaves a patient pool of the mild to moderately depressed -- and mild to moderate depression is a disease that waxes and wanes. "So the alternative explanation to the one Kirsch is offering is that these were people who weren't all that ill to begin with," says Hyman.
But even if Kirsch's case proves too extreme, that still leaves a placebo response rate of between 30 to 40 percent, meaning that 30 to 40 percent of the depressed subjects in clinical trials feel happier while taking a contentless pill. And there is good reason to think that that may be an underestimate. In the first place, many randomized controlled trials are preceded by a so-called washout phase in which all participants take an inert pill and anyone who reacts favorably to it is eliminated; the 30 to 40 percent comes out of a group, then, that has already been purged of probable placebo-reactors. In the second place, several studies show that the rate may be more like 50 percent for depressions that have lasted less than three months, especially if they were also triggered by a specific event. Given that active drugs often have unwelcome side effects (though, oddly, placebo takers sometimes report these, too) and that many patients might be relieved to know they don't actually need a chemical antidote to their woes, why shouldn't doctors give Brown's suggestion a try?
Doctors who deliberately prescribe placebos and patients who accept them are not unheard of, even now. Many of us, for example, enter into a tacit agreement to take a placebo when we ask for, and usually get, a prescription for antibiotics to treat a viral infection. (And since this is a dangerous placebo -- it contributes to the problem of bacterial resistance -- maybe it would be better, as Brown suggests, for doctors to "prescribe," as in write down, a suggestion to take an over-the-counter cold remedy. That way, they could assuage the common longing to leave a doctor's office with something in hand without contributing to a public health threat.) Moreover, plenty of Americans cherish the conviction that the body works in ways unknown and unknowable by Western medicine. "Health professionals and people of a scientific bent underestimate how strongly many people believe that the body heals itself and that synthetic drugs are bad for you and yada yada yada," as Brown puts it. "I don't think it would be as tricky as it sounds to tell somebody, 'I can give you this stuff, we don't know exactly how this works but one thing it probably does is to allow the body to generate its own healing function."'
One patient I spoke with, a man who has been participating in a study of an anti-anxiety drug, found nothing particularly objectionable about the prospect of being treated with placebo, a concept he seemed thoroughly to understand. The man, a 60-year-old real-estate agent who didn't want his name used, said he had "been kind of hyper" for as long as he could remember. He was the kind of guy who couldn't stand hitting a bunch of traffic lights in a row, as he put it, and he'd had stomach problems since he was a kid, for which he dosed himself daily with over-the-counter antacids or baking soda and water. The study he's involved in has an unusual design that requires everybody to take the active drug for the first six weeks, and this man is convinced that he has benefited from it. He feels calmer, and his stomach hasn't been bothering him at all. But next week, when the study enters a new phase, he stands a 50-50 chance of getting only a placebo, and that doesn't worry him, either. "If a placebo works, that's fine with me," he says. "The important thing is that it works, not how it works. I'd find it kind of interesting if it did, to tell you the truth. That would be a sign to me that I need to work on my mind a little more."
And there is one small study, the only one of its kind, that suggests that a placebo prescribed openly could win over some patients. At an outpatient psychiatric clinic in 1965, two researchers from John Hopkins University gave 15 adult "neurotics" an inert pill identified as such. With straight faces, the doctors told them that "many people with your kind of condition have also been helped by what are sometimes called sugar pills, and we feel that a so-called sugar pill may help you, too. Do you know what a sugar pill is? A sugar pill is a pill with no medicine in it at all. I think this pill will help you as it has helped so many others. Are you willing to try this pill?" Fourteen of the patients were convinced by this vaguely smarmy-sounding pitch (the 15th dropped out after her husband made fun of the idea), and after a week all reported ameliorated symptoms. Some thought the pill definitely was a placebo and some thought it must actually be an active drug, but either way, they had faith that the doctor was trying to help them, and they improved. At the least, then, Brown's idea deserves to be tested out with a bigger, more reliable study.
Yet in the end it is also a bit of a gimmick. If you get too fixated on the ritual of swallowing a pill, you miss the larger meaning of the placebo effect. And the larger meaning has to do with a certain kind of empathic attention that a doctor -- some doctors -- give to patients. It has to do with faith and hope and a physician's capacity for marshaling those sentiments in the service of the sick. "The secret of the care of the patient," wrote Dr. Francis W. Peabody in a popular essay for doctors, "is in caring for the patient." It may also be the secret of the placebo.
Karin Scheetz doesn't have the luxury of thinking all that philosophically about the placebo effect. But she thinks about it anyway, because it's just so weird. Scheetz is a clinical-research coordinator at a company near Baltimore called Clinical Insights, which carries out randomized controlled studies of new psychotropic drugs at the behest of various pharmaceutical makers. Her job is to deal with placebos in their most literal, not to say, banal form -- she hands them out to patients, and she makes sure the patients take them, and she notices that the dummy pills the companies provide always look and smell and taste just like the real ones (there is no sugar in them anymore), and that neither kind has any telltale writing on them.
Scheetz, 27, is a talker -- a naturally outgoing, effervescent sort. That very aspect of her personality and how it might shape the mood of an anxious person is something she thinks about a lot. That's because, right now, she is working on a study of an anti-anxiety drug for which placebo response is a real problem. "The company has tested it already," she explains as we sit chatting one morning at the Clinical Insights offices, a standard-issue medical suite in a beige-and-smoked-glass office building above a busy commuter highway. "They just haven't been able to pull out the actual drug effect, but they're sure it works. This is sort of their last chance to show it."
The company sponsors have given Scheetz particularly detailed instructions: they want her to record the amount of time she and the principal investigator, a psychiatrist named Lawrence Adler, spend with each patient. They know that the weekly sessions during which she and Adler check up on the patients, listening to their complaints and so on, can themselves be therapeutic -- and therefore confound the data on the drug's effectiveness. "I tend to be a very friendly person," says Scheetz, who has a tumble of brown curls and a broad smile. "And I've had to monitor myself, to hold back a bit. I try to be as businesslike as possible. I cut down on the additional conversation and I do an assessment in 15 minutes where normally I might take half an hour."
The placebo effect is so huge - between 35 and 75 percent of patients benefit from taking dummy pills in studies of new drugs -- that it should probably be put to use in clinical practice.
This isn't the first time a company has issued instructions aimed at controlling emotional atmosphere. Sometimes the request is for Dr. Adler to wear his lab coat in order to increase the sense of distance between him and the patients. But it's the first time they've touched on something so basic to Scheetz: her ability to communicate with people and put them at their ease.
What Karin Scheetz has stumbled upon, what is for her something of an occupational hazard, is actually an important idea. Placebos may work because they create the pretext for a doctor or a doctor surrogate to listen carefully to our troubles and to pay us a close and committed and hopeful attention. Since researchers on clinical studies are so invested in retaining participants, they are usually generous with their time and care and enthusiastic about the treatment being tested. Placebos, then, are just the tokens of a faith that somebody is at last in league with us against our illness.
For years, scientists have tried to come up with a plausible explanation for the placebo effect, and there are by now several contenders. One is classical conditioning: people who have experienced relief in medical settings or from ingesting a pill are primed, like Pavlov's dogs, to do the same again. Another is the release of endorphins: several studies have suggested that placebo pain relievers, at least, work by stimulating the brain's own analgesics. Still a third is that taking a placebo, especially if it is administered in an atmosphere of hope, relieves stress, which tends to aggravate the symptoms of placebo-sensitive conditions like asthma and hypertension.
What all of these explanations have in common -- for they are certainly not mutually exclusive -- is the element of expectation, the promise of help on the way that can only be imparted by another human being. This kind of hope seems particularly relevant when you respect the distinction made by Dr. Howard M. Spiro, one of the most thoughtful writers on the placebo, between illness and disease: illness is what the patient feels; disease is what the doctor finds. Hope can help soften the experience of illness, though it cannot cure the underlying disease. Says Walter Brown: "What we're really talking about with the placebo effect is the reduction of distress, the reassurance from being in a healing situation that you're finally in somebody's hands."
In fact, there is hard data to support the notion that the emotional alliance between a doctor and patient is itself a therapeutic force -- that a compassionate and optimistic physician can be, as one writer puts it, a walking placebo, that words can be physically comforting. In 1987, a British doctor named K.B. Thomas tried a little experiment on 200 patients in his own practice who came to see him feeling under the weather, but with no abnormal physical signs. Thomas gave one group of patients a definite diagnosis and told them they'd be better in a few days. He told the other that he couldn't be sure what was the matter with them and so couldn't say when their symptoms would clear up. Two weeks later, 64 percent of the patients who received an encouraging consultation had gotten better, compared with 39 percent in the other group.
In a study on postoperative pain conducted at Massachusetts General Hospital, half of the patients were randomly chosen to be visited by the anesthetist the night before surgery and treated, poor dears, in a brusque, offhand manner. "My name is so-and-so," the doctor would say. "Tomorrow I'm going to give you anesthesia. Don't worry; everything is going to be all right," and then off he'd go. The same anesthetist visited the other group of patients and this time put on a warm and sympathetic face, holding their hands as he sat on their beds, telling them exactly what to expect in the way of postop pain and assuring them that pain relief would be available. Those in the second group ended up requiring only half the amount of painkilling medication and were discharged an average of 2.6 days earlier.
A practitioner's enthusiasm for a given treatment and his ability to convey it to patients also seems to make a difference, whether the treatment is active or a placebo. "I think it may be the key factor," says Daniel Moerman, a medical anthropologist at the University of Michigan at Dearborn. "If a physician thinks this new drug is the greatest thing since sliced bread or if he's really excited about a new theory, it always makes a difference. That's one reason why the effectiveness of older drugs often wanes when a new one comes along."
Accept all this, though -- that the placebo effect is real and that it probably works through a certain kind of expectation generated by empathic care -- and you are brought face to face with a quandary. For many of the elements of the doctor-patient relationship most likely to produce a placebo effect are in decline these days.
Time, for instance. In 1943, the average visit to the doctor lasted 26 minutes; in 1985 it lasted 17. Under managed care, the incentives to jack up productivity -- which is to say, the number of patients seen in a day -- are only growing. In a 1995 survey of more than 1,700 doctors, 40 percent said they were spending less time with patients than they had even three years earlier. "A large number of the depressed patients who enroll in our new drug studies do so because they're unhappy with the care they're getting through their H.M.O.'s," says Dr. Angelo Sambunaris, the chief clinical investigator at an Atlanta drug-testing company. "They'll say, 'I've seen more of the staff here in two weeks than I did in two years at my H.M.O.,' or 'Here you start me on a new medication and say come back in a week. My own doctor says call back in a month."'
With employers constantly changing health plans, long-term relationships between doctors and patients have dwindled, to the detriment of the placebo effect. "Clearly, one of the economic factors that undermine the placebo response is changing doctors a lot," says Dr. Howard Brody, the professor at Michigan State. "It undermines the kind of solid relationship that builds trust and a sense that your doctor understands your story."
Moreover, many doctors have shed some of the old-fashioned professional baggage that once transformed them into walking placebos. The availability of high-tech diagnostic tools and the shrinking of appointment slots means doctors are doing fewer physical exams, even though the laying on of hands is perhaps the most basic source of human comfort and the annual physical one of the tried and true bonding rituals between doctor and patient. Sensitive to charges of paternalism or even malpractice, many doctors are likely to hit the cautionary note harder than the hopeful one when they offer a prognosis, and to convey more ambiguous information and less direct advice than patients might actually want. "A colleague of mine told me that he was on grand rounds with a physician who clearly had a strong recommendation about what would be best for a patient, but didn't tell the patient," says Brody. "My colleague asked him why, and he said, 'Oh I couldn't, the patient has rights.' But of course, there's a difference between coercing people and making recommendations. Patients have a right to our recommendations and we have a duty to share them." However savvy Internet-era patients may be, however indignant they might be at the thought of being patronized, many would still like reassurance from their doctors -- and clear guidance too.
Sometimes our contemporary notions of informed consent, important though they are, interfere with a doctor's capacity to offer hope and comfort. "When, for example, a patient who is about to undergo a straightforward surgical procedure anxiously asks the surgeon, 'Doctor, am I going to make it?' what should the surgeon say?" asks Walter Brown. "Traditionally, the response would be to encourage the patient: 'You're in good hands here, everything will be fine.' Yet I have posed this scenario to a variety of physicians, both residents and attendings, and found that some consider themselves obligated to list the possible complications and the risk of death, even if the mortality rate from the procedure is minuscule."
Medical training that (rightly) values precise expression and a culture that values full disclosure and protection from liability may spur doctors to tell their patients the uninflected and deflating truth even in the most dire circumstances. But surely telling the whole truth and nothing but the truth is not the most humane or practical of counsels for doctors. Someone I know of, a physician herself, was recently operated on for a brain tumor. Her doctor chose his first postop visit to her in the hospital to tell her she would never work again. It didn't help her recovery much.
Meanwhile, doctors demoralized by H.M.O. bureaucracy may be less able to convey the aura of authority that soothed generations of patients. In the same study in which doctors reported spending less time with their patients, a third also said they were somewhat or very dissatisfied with medical practice. Another 1995 study of 2,000 physicians across the country found that in regions where managed care predominated, doctors complained of pressure to see more patients and bemoaned restrictions on their ability to order tests and prescribe drugs. "Physicians' time is increasingly consumed by paperwork that they view as intrusive and valueless," The New England Journal of Medicine editorialized recently.
Of course, physicians are not entirely helpless. "Someone wrote in a letter to The New England Journal saying, 'Who's preventing us from spending time with our patients?"' says Brody. "And he had a point. He said: 'Look in your managed-care contract. Does it say you can only spend five minutes with your patients? No, it says if you want to make as much money as you did last year, then you have to see patients at certain rates, so that you can meet quotas for reimbursement.' And that's a little different. As doctors, there are things we can do." Yet surely it matters that, as the Journal put it, "many physicians are dismayed. Some are frankly morose." Who wants a morose doctor? And how could he ever inspire the hope that underlies the placebo response?
In 1980, The New England Journal of Medicine published an eloquent essay by its former editor Franz Ingelfinger, in which he argued that though arrogance was usually a pejorative word, doctors could exercise it in a way that was humane and salutary. If arrogance meant "insolence, vanity, arbitrariness or lack of empathy," it was worthless or worse to a patient. But if it meant the confidence some physicians conveyed in their own advice, then it was a gift to the sick, who are often frightened and, in their fear, a little childlike. Ingelfinger came to this belief when he himself was ill. He had developed cancer of the esophagus, the very organ he had spent much of his professional life studying, and had to decide whether to undergo chemotherapy or radiotherapy. From his friends and colleagues across the country, contradictory suggestions poured in. "As a result," Ingelfinger recalled, "not only I but my wife, my son and daughter-in-law (both doctors), and other family members became increasingly confused and emotionally distraught. Finally, when the pangs of indecision had become nearly intolerable, one wise physician friend said, 'What you need is a doctor.' He was telling me to forget the information I already had and the information I was receiving from many quarters, and to seek instead a person who would dominate, who would tell me what to do." When he found that person, Ingelfinger felt an "immediate and immense relief."
Is it possible to stimulate a placebo response without dispensing a placebo? Almost certainly. Intuitive -- and benignly arrogant -- doctors have been doing it for years. And why shouldn't they? After all, placebo therapy, if you want to call it that, is safe, cheap and particularly useful for the "worried well" -- patients who feel ill but have no organic disease and who account for so many visits to the doctor's office. "If you can create a more effective healing practice based on the manipulation of meaning, then you're going to heal more people," says Daniel Moerman. "Is it some sort of failure if it isn't due to a pill? The important thing is you've made somebody better."
Besides, the physician who can marshal a placebo response with her words and manner probably comes closest to what many of us would think of as the profession's ideal -- the kind of doctor who seems wholly committed to our welfare, not the insurance company's; who knows when and how to give us hope, who listens closely but doesn't feel constrained from delivering advice; who knows us because she has taken the time to know us.
"The placebo effect can occur," as the physician Herbert Spiegel once put it, "when conditions are optimal for hope, faith, trust and love." It might sound sentimental, but then sentiment, working hand in hand with science, can make medical practice so much more powerful. A world in which placebo -- preferably in the form of deft encouragement, but sometimes in the form of a harmless pill -- was tolerated, even embraced, would be a world in which doctors never forgot that medical practice consists not only of the technologies of diagnosis and treatment but also of the careful tending of a patient's expectations and the unabashed willingness to comfort. It would be a world in which doctors kept in mind the uses of optimism, knowing that there is, as Brown points out, "a great difference between telling a patient, 'This is a very powerful painkiller,' and saying, 'This might help your pain a bit"' -- and that they might as well give the former speech if they're going to bother prescribing medicine at all. It would be a world in which many of us might feel a lot better.
Return to ChiroZINE ARCHIVES
Return to the PROBLEMS WITH PLACEBOS Page