In “A Duty to Deceive: Placebos in Clinical Practice,” Bennett Foddy argues that physicians should use placebos in certain circumstances. He makes some fine points, but I didn’t buy every step of his logic.
According to the American Medical Association’s (AMA’s) code of medical ethics, a placebo may be used “only if the patient is informed of and agrees to its use.” Since a placebo’s effectiveness hinges on believing it will work, which for most people means believing it includes some active agent or procedural step (in the case of surgery), disclosing that a pill or a procedure is a placebo is likely to render it useless.
But the AMA’s guidance offers a back door. It continues,
A placebo may still be effective if the patient knows it will be used but cannot identify it and does not know the precise timing of its use. A physician should enlist the patient’s cooperation by explaining that a better understanding of the medical condition could be achieved by evaluating the effects of different medications, including the placebo. The physician need neither identify the placebo nor seek specific consent before its administration.
Perhaps whether or not you think this is ethical depends on whether you think it qualifies as informed consent and whether you think informed consent is necessary anyway. (Though, if not, an even greater level of subterfuge may be acceptable to you.)
Foddy makes an argument for a greater degree of deception in use of placebos than the AMA recommends, but only when no other therapy is superior.
[W]hen placebos are used in good faith by a doctor who believes no better treatment to be available, it becomes possible to alleviate symptoms that cannot otherwise be alleviated. Since deceptive placebos are sometimes the best treatment, it is possible that they may be prescribed in a manner that is ethically defensible.
Further, Foddy thinks the AMA has gone too far.
If patient expectations can control the magnitude of the placebo effect, then the AMA is just wrong when it suggests that doctors can reveal the inert nature of placebo treatments without compromising their efficacy. If placebos are to be used in the clinic at all, they ought to be used deceptively.
It’s hard for us to become comfortable with deception, as it challenges autonomy. And yet, if a (deceptive) placebo is used only when it’s the best available treatment, perhaps we should welcome it. Foddy has some advice for patients.
I am not sure that patients have an expectation of complete honesty from their physicians (at least in the absence of AMA provisions that explicitly mandate it). But even if patients hold this expectation, they ought not to hold it with regard to the deceptive use of placebos. Responsible placebo use, being unlike other forms of clinical deception, bears no risk of making patients worse off, beyond the baseline risks incurred by any other effective therapy. It is a type of deception that patients ought to be thankful for, just as we are thankful when we receive a mendacious compliment from a friend.
I disagree a bit here. Some placebos come with risks that could make patients worse off. Consider sham surgery that would require anesthesia (which carries risks) and perhaps some penetration of the skin. Even sham acupuncture penetrates the skin. Surely there’s some risk of infection. But there’s yet another consideration, which is suggested by the following passage in which Foddy argues that informed consent is not as vital in clinical practice as for clinical research.
Research and clinical practice are not alike. In research, the subjects provide a service (data) to the researchers. If the researchers do anything to the subjects without their express consent, then they are ignoring the wishes of the subjects and treating them as mere means to obtaining the data. By contrast, in the case of clinical practice, a patient can always refuse treatment, even if the treatment is a deceptive placebo. It is thus impossible for a doctor to instrumentalize a patient who presents of her own free will to the clinic, as long as some beneficial treatment is offered in exchange for payment.
First of all, this is illogical because a subject in a study can also refuse treatment or switch form control to treatment groups in many cases. Refusal and trading information for participation are two different things. In addition, though Foddy mentions the fact that in clinical practice doctors exchange provision of treatments for money (as well, in many cases, information, though not always formally), he doesn’t connect the dots. Payment to the doctor is a cost to the patient, society, or both, depending on financing arrangements. In other words, receipt of a placebo is never completely costless. Even the time spent receiving it is a cost.
One could argue that patients can and should make a cost-effectiveness evaluation, that we need not interfere with a placebo-for-hire transaction on their behalf. Foddy did not make this argument. But even if he had, it’s problematic because patients rarely pay the full cost at time of treatment. Placebos, therefore, might be over-provided relative to their value.
In addition, the employment of deception necessitates—indeed is—an information asymmetry, a type of market failure. Is the doctor to tell the patient not only that a placebo will help but how much it will do so (e.g., 50% reduction in pain or similar)? To maximize effectiveness, shouldn’t the doctor overstate its likely amount? In effect, isn’t all placebo prescribing just this? Doesn’t this interfere with the patient’s ability to make a rational cost-effectiveness assessment?
None of this necessarily means it’s wrong to provide placebos or use deceit in doing so. My point is that the bar over which one must pass to justify these is never at ground level. There is both an ethical bar, which is Foddy’s concern, and a cost-effectiveness one, which he ignores but shouldn’t.