One patient stands out in the memory of Stephen Straus, M.D., for her remarkable recovery, more than 10 years ago, from chronic fatigue syndrome. The woman, then in her 30s, was "very significantly impaired," says Straus, chief of the Laboratory of Clinical Investigation at the National Institute of Allergy and Infectious Diseases. "She had no energy, couldn't work, and spent most of her time at home." But her strength was restored during a study to test the effectiveness of an experimental chronic fatigue drug.
"She and her parents were so thrilled with her recovery that they were blessing me and my colleagues," recalls Straus, the principal investigator on that study.
Like many drug studies, the chronic fatigue medication trial was a "placebo-controlled" study, meaning that a portion of the patients took the experimental drug, while others took look-alike pills with no active ingredient, with neither researchers nor patients knowing which patients were getting which.
It's human nature, Straus explains, for patients and investigators alike to try and guess in each case: Is it the real drug or a dummy pill? But people shouldn't kid themselves, he says, that they can consistently tell the actual drug from the sham by seeking out tell-tale signs of improvement.
Turns out, the woman's quick turnaround from chronic fatigue occurred after taking placebo pills, not the experimental drug. Straus says, "She was amazed by the revelation that she'd gotten better on placebo."
Research has confirmed that a fake treatment, made from an inactive substance like sugar, distilled water, or saline solution, can have a "placebo effect" -- that is, the sham medication can sometimes improve a patient's condition simply because the person has the expectation that it will be helpful. For a given medical condition, it's not unusual for one-third of patients to feel better in response to treatment with placebo.
"Expectation is a powerful thing," says Robert DeLap, M.D., head of one of the Food and Drug Administration's Offices of Drug Evaluation. "The more you believe you're going to benefit from a treatment, the more likely it is that you will experience a benefit."
To separate out this power of positive thinking and some other variables from a drug's true medical benefits, companies seeking FDA approval of a new treatment often use placebo-controlled drug studies. If patients on the new drug fare significantly better than those taking placebo, the study helps support the conclusion that the medicine is effective.
Benefiting from Belief
Researchers have been studying the placebo effect for decades. In 1955, researcher H.K. Beecher published his groundbreaking paper "The Powerful Placebo," in which he concluded that, across the 26 studies he analyzed, an average of 32 percent of patients responded to placebo. In the 1960s, breakthrough studies showed the potential physiological effects of dummy pills -- they tended to speed up pulse rate, increase blood pressure, and improve reaction speeds, for example, when participants were told they had taken a stimulant, and had the opposite physiological effects when participants were told they had taken a sleep-producing drug.
Yet, even after 40 years, big questions remain about the interplay of psychological and physiological mechanisms that contribute to the placebo effect. Today's brain imagery techniques do lend support, though, to the theory that thoughts and beliefs not only affect one's psychological state, but also cause the body to undergo actual biological changes.
The phenomenon needn't baffle us, says Michael Jospe, a professor at the California School of Professional Psychology who has studied the placebo effect for more than 20 years. He points out that all people experience physiological reactions to anticipation and stress -- something like the fight-or-flight response -- that help them to survive and cope. When you step out of your office and a spider jumps out at you, Jospe analogizes, "you'll get a fright and have a physiological reaction. And the next time you go out that way, the thought that it could happen again can produce a physiological reaction before you even open the door." So, he says, the relationship between a thought and a negative psychophysiological reaction like fear is something we experience daily.
That goes for positive associations, too, Jospe continues. "The placebo effect is part of the human potential to react positively to a healer. You can reduce a patient's distress by doing something which might not be medically effective." It's like kids and Band-Aids, Jospe says. "When you put a Band-Aid on a child and it has stars or comics on it, it can actually make the kid feel better by its soothing effect, though there's no medical reason it should make the child feel better."
There is no medical reason, either, that look-alike placebo tablets used in a 1997 study of benign enlargement of the prostate gland should have made the study participants feel better. But in this Canadian study, more than half of the men who got the placebo pills reported significant relief from their symptoms, including faster urine flow. Researcher J. Curtis Nickel theorized that the patients' positive expectations of the experimental drug's benefits may have caused therapeutic smooth muscle relaxation by decreasing nerve activity affecting the bladder, prostate and urethra. Study participants on placebo complained of side effects, too (sometimes called the "nocebo" effect), ranging from impotence and reduced sex drive to nausea, diarrhea and constipation.
It's this powerful placebo effect, coupled with the fact that many medical conditions involve a natural course of better and worse periods (arthritis and multiple sclerosis are examples of diseases with flair-ups and lulls), that can make it difficult to know if a health upswing should be credited to a drug effect. One way to account for such variables in a drug study: give one group of patients placebo and another the experimental drug, and see if the drug group's health improvements sufficiently surpass those from placebo. In Straus' study, the chronic fatigue syndrome drug failed to adequately demonstrate its superiority over dummy pills.
Proof in the Placebo
FDA doesn't require that a drug study include a placebo control group, DeLap says, only that its design be capable of establishing a drug's safety and effectiveness. Non-placebo types of drug studies include "head-to-head" studies, which compare the experimental drug to an existing treatment, and historically controlled studies, which compare the new drug's effects with information gathered in the past about the expected progression of a medical condition.
Often, however, a placebo control can provide the clearest insight into what a treatment can accomplish, according to DeLap, especially with some psychiatric and other drugs in which the placebo effect is known to play a particularly weighty role. In fact, DeLap says, in some cases the placebo effect "makes it almost hopeless, statistically" to use studies that test a new treatment side-by-side against an existing one and determine whether the new treatment works.
The placebo controls that have traditionally been used to test medications have recently been used, too, to test the effectiveness of surgical procedures. In one well-publicized study sponsored by the National Institutes of Health, half of the Parkinson's disease patients enrolled in the trial underwent a sham surgery in which doctors drilled holes into their skulls but didn't implant the potentially beneficial human fetal tissue in their brains.
While FDA doesn't evaluate the safety or effectiveness of most surgical techniques, the agency does regulate surgical implantation of animal cells or re-engineered human tissues. The agency has approved at least one sham surgery-controlled trial, which will study the effectiveness of implanted pig fetal cells for Parkinson's.
Even with the powerful scientific advantages of including a placebo control, researchers and FDA must look at each treatment individually to decide if the use of placebos is appropriate and ethical. In fact, much medical research does not involve a placebo control because "it's just not an option, ethically," DeLap says.
To determine whether a placebo-controlled trial is acceptable, drug company and FDA experts ask: For what condition is this drug being tested? What is the natural progression of the disease? How serious is the risk if a patient gets a placebo rather than an active treatment?
In DeLap's specialty, oncology, placebo-controlled studies are often unacceptable because of the great risk to cancer patients of any treatment delays. For a headache, on the other hand, patients in a study may be uncomfortable for a time, but are not at risk of a lasting health impact. So, for those conditions in which the downside of being on placebo is modest and short-lived, DeLap says, it's an individual's prerogative to say, "I know what I'm getting into, and I want to further this scientific research."
To help ensure that patients know the pros and cons of enrolling in a study, each participant must sign an "informed consent" form, which clearly explains:
the purpose of the study,
what enrollees will be asked to do (take a pill twice a day for three months, for example, and visit the doctor once a week for blood and other laboratory tests),
the possible benefits and known adverse reactions associated with the experimental treatment, and
other therapies that are available for their condition.
Even willing participants can't sign away their right to a well-designed study, though, DeLap points out. "We can't fall into the trap of thinking that, once someone says 'I'm willing to participate,' their consent covers us for deficiencies. Our responsibilities go way beyond getting informed consent." One such responsibility: As a study progresses, researchers monitor results so if major positive or negative drug effects are seen, the study can be stopped. The first major clinical study of the AIDS drug AZT (zidovudine), for example, was halted early when researchers saw that AZT patients were living significantly longer than others in the study.
Still, some critics say today's safeguards are not sufficient and oppose the use of placebo-controlled studies in almost all drug research. Patients desperately seeking to end their suffering, some say, may not be capable of giving true informed consent.
DeLap and other FDA experts view any far-reaching ban on placebos in research as paternalistic. "We at FDA don't have an ethical blind spot, as some would suggest," DeLap says. "A patient's right to the best treatment is always paramount. But the social hope is that careful scientific research can help us learn beyond a shadow of a doubt what works and what doesn't, so that these patients' kids will have better treatments available to them."
Tamar Nordenberg is a staff writer for FDA Consumer.
Placebos in the Doctor's Office: Opposing ViewsToday, when 26-year-old Jennifer Kennedy (not her real name) from Rockville, Md., has a panic attack, she simply distracts herself with a book or crossword puzzle. But at their height, several years ago, the attacks scared her so much that she finally went to the emergency room. "I thought I was going to die from a heart attack," says Kennedy.
She didn't believe her doctor when he told her to "just relax" and she would feel okay, and he ended up prescribing pills. He didn't tell her until two weeks later, after her panic attacks had subsided, that the pills were simply placebos with no active medical ingredient.
Those who oppose the use of placebo pills in medical practice say that such deceit can undermine the essential trust between patient and doctor. Gastroenterologist Michael Kirsch, M.D., has called doctors who prescribe placebos outside medical research "con artists." Kirsch asks in a 1998 editorial in Priorities, the magazine of the nonprofit American Council on Science and Health, "If using placebos therapeutically is ethical and reasonable, shouldn't we encourage judges to render extralegal activist rulings, winegrowers to bottle 'placebo' vintages, curators to display masterpiece look-alikes misleadingly, and journalists and newscasters to sanitize news?" He continues, "In such a world, all of us would be groping for truth in a hall of mirrors."
But placebo researcher Michael Jospe disagrees with what he calls this "strict, grumpy approach that concludes that any doctor who uses placebos is acting unethically." Some circumstances, he says, justify this kind of benevolent deception -- like when a patient insists on a medicine that is unnecessary and carries needless risks.
"You've got to be there on the oncology ward," Jospe says, "and see how suffering people get so demanding of drugs that might be extremely harmful to them. If you look at sugar pills in the broader context of a supportive doctor-patient relationship rather than just as ripping off the patient, you may come to a different conclusion" about the ethics of placebos.
It's not uncommon for a patient to feel betrayed initially upon hearing that they were given a placebo, Jospe says, but a sensitive doctor can explain to the patient, "No, that the placebo worked doesn't mean you're crazy. You were just in distress and thus more prone to reacting to anything with the potential to help."
Kennedy admits to feeling deceived when she first found out the pills her doctor prescribed were fake, but says she now appreciates the doctor's decision to prescribe the placebo. "At first, I felt stupid. But that day at the hospital, the doctor must have realized I wasn't going to accept 'you're fine, it's all in your mind.' The placebo helped me realize that I'm not unhealthy and I'm going to be okay. Now, I think it's really neat that something that didn't really affect me had such a big effect on my life."