A recent study published in the Journal of General Internal Medicine and featured in a Time Magazine article, indicated that of 466 academic physicians in the Chicago area, 45% indicated that they have prescribed a placebo for a patient. This has sparked a discussion of the ethics of prescribing placebos in particular and deception in general in medicine.
A placebo is a biologically inactive treatment, such as a sugar pill. Any perceived benefit from taking a placebo is due to a combination of factors, mostly biased observation and non-specific effects, collectively referred to as the placebo effect. I discussed the placebo effect at length last week, and now will delve deeper into the question of deception in medicine more generally.
Prior to about 30 years ago the relationship between a physician and their patient was functionally paternalistic. This means that the physician did what they thought was best for their patients as a parent would for their child. It also meant that “benign deception” was often used, including prescribing treatments that were known to be inactive or ineffective. Sometimes the deception was one of omission – for example, not telling a patient that their disease was terminal and incurable so as not to upset them needlessly.
An illustrative example of this type of deception was the practice of prescribing methylene blue to patients as an active placebo. This substance is quickly filtered unaltered by the kidneys and will turn the urine deep blue. I was told first hand by older physicians of the practice of telling patients that their condition was caused by a benign tumor, and that the medicine (methylene blue) would entirely dissolve the tumor. The patients were told that they would know the treatment has worked when their urine turns blue. The purpose of this was to convince a patient thought to be a hypochondriac, or to have psychologically-based symptoms, that their fake ailment was treated, in the hopes that they would take comfort from this and their symptoms would resolve.
It is easy from our modern perspective to be judgmental of the paternalism of past generations, but I feel that it is proper to judge people in their own historical context. The fact that such practices seem so outrageous to modern sensibilities is not necessarily a marker of modern moral superiority but may simply reflect the general changes in American society in the last half-century.
The Physician-Patient Relationship Today
The more accepted model of the physician-patient relationship today (or health care provider-client, to be more general) is that of cooperation, with the patient as a full partner. This is based upon the principle that everyone has a right to be fully informed so that they can make the best decisions for themselves. Further, individuals generally wish to be more empowered to control their own health care. This trend has also been driven by the easy accessibility of medical information to the general public. It is not uncommon now for patients to present to their doctor with a hand full of pages printed from the internet.
In reality there is still a broad spectrum from old-fashioned paternalism to the modern cooperative model. This is partly regional – in the US the coasts tend to follow a more cooperative model, while in the mid west and a lesser degree the south still retain a degree of paternalism. It is therefore not a surprise that the above referenced study was conducted in the mid west, and I suspect that the numbers would have been different if conducted in the northeast.
There is also a generational difference, as we would expect, with older physicians tending to have an older style of practice. Superimposed upon these trends are individual differences among physicians. And, in my experience, I think that patients themselves largely determine the nature of the relationship they have with their physicians. Even when patients are fully empowered to control their health care, some will voluntarily surrender to a greater or lesser degree to the expertise of their physicians. Physicians have to, in fact, be in tune with their patient’s desires in this regard. Put simply, some patients want information, and other patients want to be told what to do, and most want some combination of both.
In the context of the modern physician-patient relationship deception is generally not acceptable. Within the practice of medicine there are various specific situations that need to be discussed.
Medical research is the easiest situation to deal with because there are specific regulations that remove most of the guesswork. Research involving humans requires adherence to strict ethical rules and the oversight necessary to enforce those rules. The guiding principle of medical research is informed consent. Subjects need to be told exactly what the potential risks of an experimental treatment are, what (if any) are the potential benefits to them, and the probability that they are receiving a placebo treatment. They need to be told exactly what will happen to them in the study, what the treatment alternatives are, what will happen if they have a negative side effect, and they need to be given the opportunity to ask any questions to their satisfaction.
Informed consent also applies to the practice of medicine outside of research. This means that patients have a right to similar information and to have any questions answered prior to agreeing to have surgery or receive a drug or treatment. However, the complexity here is that there is really no such thing as “full” disclosure. It is not practical for a physician to go over every possible consequence of a treatment prior to prescribing or initiating it. Potential side effects that are benign and/or extremely rare do not necessarily need to be discussed directly. Physicians do exercise judgment in deciding what patients need to be told.
Often, after the important facts about a new drug are discussed, patients may be referred to the package insert, which has the full prescribing information approved by the FDA. This way the patient can have as much information as they desire about a new drug, and then can ask their physician about anything that concerns them.
Keep in mind that I am not saying that what I described above is what always happens, but it is generally accepted to be good practice. Also, physicians that neglect to follow such good practices open themselves to liability if a negative outcome does occur.
Another context which is fairly straightforward is the need for full disclosure. This means that physicians are ethically required to tell patients any relevant information. In addition to having full consent prior to a treatment, patients are considered to own their health care information and have a right to access to it. Likewise, physicians have a duty to inform patients about anything relevant, such as test results or their diagnosis or prognosis.
I find this to be the most fuzzy area, however. Certainly, factual information needs to be given to patients. However, facts do not always tell the whole story, they must be put into a clinical context. Therefore, for example, a physician may emphasize the possibility for hope or they may downplay unrealistic hope in order to focus the patient on making practical decisions. Neither approach is objectively better – they represent trade offs, a balance between keeping the patient’s outlook positive to avoid depression or nihilism, while still giving them a realistic view of their condition so they can make rational decisions.
Also, when a patient is facing a difficult health decision, for example whether or not to have a risky but needed surgical procedure, the facts can often be used to support either decision. A physician may believe, and the facts may support the conclusion, that the patient’s best chance to avoid a bad outcome is to have surgery. The physician may therefore emphasize the risks of not getting the surgery. Exactly how much time and effort is spent going over the possible negative outcomes if a condition is not treated can affect the patient’s decision. In the end the physician’s clinical judgment determines what is proper disclosure to the patient.
On the other end of the spectrum, a physician may emphasize to a patient that there is ongoing research into their disease that may produce a treatment. Is this giving the patient false hope, or is this just good bedside manner?
A very common context in which the need for full disclosure comes up is when a physician believes that a patient’s symptoms are entirely psychiatric. (For the sake of this discussion I am considering situations in which the symptoms are actually psychological, not a misdiagnosis.) A brutally honest opinion given to the patient may sound something like this: “I believe that you do not have any real physical ailment. The symptoms you are having are purely psychological – they are nothing more than a delusion on your part. Therefore the real underlying problem is a mental illness, or perhaps simply a personality disorder. These are notoriously difficult to treat, but you may benefit from counseling or psychiatric treatment.”
I have, in fact, heard a colleague say almost exactly that to a patient – with perhaps just a bit more tact. There is evidence, in fact, that in certain situations brutal honesty, while uncomfortable for the physician and the patient, may result in the best outcome. The patient may protest, but then the symptoms will stop.
However, such honesty will often sacrifice the therapeutic relationship, so in most situations physicians will take a gentler approach. For example, I might state that certain symptoms are known to be caused by stress and that they tend to resolve on their own. This is completely true, but has a different tone and emphasis than the brutally honest statement above.
What about the outright deception of prescribing a known placebo to a patient? Today this is generally considered to be unethical, not only because it prevents informed consent but because it interferes with the trust that is a necessary prerequisite for the physician-patient therapeutic relationship. I agree with this position and would never prescribe a placebo or directly deceive a patient.
However, again there is a gray area. My perspective is biased in that I am a neurologist, so many of the medications I prescribe have multiple uses. A pain medication may also be an antidepressant, or a muscle relaxant may also be a sleep aid. I therefore might recommend a specific medication to a patient for one effect but partly base my choice on a secondary effect that I think will also be helpful. I may emphasize to the patient the effect they are more likely to understand and accept and will not necessarily get into all the details that led me to that specific choice. But if a patient asks me a direct question, I always give them a straight and truthful answer.
Another gray area is that physicians may be very optimistic with their patients about the potential of a new treatment in order to maximize (or so they think) the placebo effect. If a physician prescribes a legitimate pain killer for an appropriate situation and tells the patient “I think this is really going to help with your pain,” when a more accurate statement might be “there is about a 60% chance this will help with your pain” is that subtle deception justified on the grounds that it may enhance a placebo benefit from the medication? (As an aside, I am not convinced that it will, but that is a separate question.)
The same is true of negative side effects. Some physicians worry that if they go over every possible little side effect that patients will get side effects just because they are worrying about it so much or because they become hypervigilant to any slight symptom. While I disclose anything important, and point them to the package insert for full information, I may say truthfully to patients that most people I have prescribed this drug report no side effects. Again – there is wiggle room in terms of emphasis and detail.
To summarize, I think that within modern ethical medicine there is no room for deliberate outright deception. Informed consent, full disclosure, and honesty are necessary within the context of medical research and a cooperative therapeutic relationship. However, there is room for good bedside manner, compassion, and for balancing hope with realism in terms of what information is emphasized, but this is a gray area.
All of this, of course, is in the context of scientific medicine. It is impossible to apply these ethical standards when dealing with a health care system that is not evidence-based. Virtually all treatments given under the umbrella of complementary and alternative medicine (CAM) or integrative medicine would be considered experimental within mainstream medicine, yet are generally given without their experimental or unproven nature being disclosed and without all the other ethical guidelines that govern mainstream medicine.
Also, many CAM treatments, like homeopathy, are nothing but placebos. Some practitioners even acknowledge this, but argue that placebos have real effects. (Again, see my prior entry for a discussion of this.) Within scientific medicine this would amount to unacceptable deception. The sectarian views of many CAM practitioners, however, often means that they believe the misinformation they give to their patients.
It is therefore difficult to impossible to have a workable system of medical ethics within an unscientific sectarian health care belief system. Some ethics can only be applied within a system that relies upon verifiable and objectively established standards. Therefore only a system based upon transparent and objective science can be truly ethical.