Is prescribing placebos an ethical practice?

February 18, 2009

placebo.jpgCommentary by Antonella Surbone, MD PhD FACP, NYU Department of Medicine, Clinical Correlations Ethics Section Editor

In October 2008, BMJ published a research article entitled “Prescribing placebo treatments: results of a national survey of US internists and rheumatologists.” Of 679 physicians who responded to a cross sectional mailed survey, over 50% reported having prescribed placebos on a regular basis. Placebos included varied from saline (3%), sugar pills (2%), OTC analgesics (41%), vitamins (38%), antibiotics (13%) and sedatives (13%). According to the authors’ definition, the placebo effect refers to positive clinical outcomes caused by a treatment that is not attributable to its known physical properties or mechanism of action, but rather to positive expectations or hopes in patients, derived from the clinical encounter. Placebo treatments, however, are quite different from each other, as saline injections or sugar pills are considered inert substances, while prescribing antibiotics or sedatives may carry adverse consequences for individual patients as well as for public health.

In the survey, physicians prescribing placebo treatments to their patients rarely used the word “placebo” (5%), in most cases describing them, instead, as treatment potentially useful, though not typically used for the patient’s illness. The authors conclude that “physicians may not be fully transparent with their patients about the use of placebos and might have mixed motivations for recommending such treatments.” Among these mixed, often conflicting motivations, the authors found the belief that a benign and safe treatment may carry some chance of a pharmacologic benefit, or might promote positive expectations in patients and reinforce their feeling of being taken care of. Some physicians treating chronic incurable illnesses may perceive the need to do something for their patients, even in the absence of any proven efficacy.

In acknowledging potential limitations of their results, the authors mention that the question items were part of a larger survey on complementary and alternative medicine and that responding physicians may have been “more favorably disposed to prescribe placebo treatments than most physicians.” Indeed, as western patients and physicians are becoming more accustomed to integrating traditional with complementary medicine, many tend to confuse placebo effects with the therapeutic ones derived from non western treatments. In oncology, for example, physicians who do not believe in alternative measures at times condescend to patients’ preferences using them in the name of a supposed placebo effect of complementary medicine The mistake inherent in such an attitude generates further confusion about the ethical justification of placebo treatments.

The study results, however, are in agreement with recent published data on the use of placebo treatments in the US and other countries, including Denmark, Sweden, Britain, Israel, New Zealand and several European countries. In my opinion, the most interesting study finding, however, is that 62% of physicians who participated in the survey believed prescribing placebos to be an ethical practice. The authors recommend further ethical and policy debate on the appropriateness of the use of placebos, yet in the article they do not explicitly argue against it on ethical grounds.

I studied medicine and initially trained in medical oncology in northern Italy, at a time when cancer patients were treated with chemotherapy without being told the truth about their diagnosis or the nature of their treatments. I recall cancer patients being at times injected with inert solutions when their pain had not responded to state-of-art analgesia at the time. The approach was deemed to be ethically justified, as the placebo effect was enough to promote a feeling of reassurance in several patients that may have taken the edge off their intense pain, by alleviating the psychological distress of feeling abandoned because their condition was helpless. Though I have witnessed many patients falling asleep after a placebo injection, and I could not feel anything but happy to see their suffering alleviated for a few hours, I always felt very uncomfortable with this practice. In my view as a young oncology fellow, giving a placebo without correctly representing it was yet another form of deception toward cancer patients, already kept in the dark with regard to their diagnosis, treatment and prognosis.

When I came to the US and started my training at the NCI, I saw that telling the truth to cancer patients didn’t hurt them psychologically. On the contrary, it gave them more inner strength to fight their illness with full awareness and supported their capacity to express their own priorities and desires and share in making important decisions. I was relieved that deception was no longer part of my practice, as I am now that international reports indicate that disclosure to cancer patients is increasingly common worldwide.

How do we reconcile truth telling with respect to patient autonomy and dignity with the prescription of placebos to patients? The AMA discourages the use of placebos by doctors who represent them as helpful, as it can undermine the trust upon which the patient-doctor relationship is based, and thus harm the patient. This seems to me a simple, clear and honest position. If we need deception in any form with our patients, it is because our therapeutic alliance with them is not solid enough.

The issue of placebos, however, is rendered more complex by the fact that placebos have been given in randomized trials with patient consent, but even this practice is still under debate. For example, the American Society of Clinical Oncology has recently issued a position statement regarding the ethical requirements for randomized study using placebos in oncology. In addition, patient advocates have stressed the importance of involving patients at the early stage of study designs involving placebos and obtaining full non-coerced patient informed consent.

The results of some studies of hypertension and depression have shown improvements in the placebo arms. Prescribing placebos in some clinical circumstances could be the right thing to do. Why, then, would we need to be less than transparent? If we believe in a placebo effect, why not our patients, in offering the placebo, that a “non-active substance” has been shown to help some patients, that we don’t yet understand whether the benefits are psychological or whether patients whose illness may sometimes remit spontaneously ascribe the improvement to the placebo.

It can be argued that the placebo effect may be invalidated by such transparency, and some physicians prescribing placebo may be afraid of undoing the placebo effect if they discuss it as such with their patients. For example, in certain psychosomatic conditions accompanied by high levels of physical discomfort for which no therapy yet exists, the lack of any treatment can generate frustration in patients who may feel judged and labeled by their physicians. Some physicians may prescribe mild or inert substances, although they may not be convinced about any therapeutic effect. Can this be effective? Probably yes, in some cases. However, is a pill the right answer to existential or psychological suffering? Are medications the right substitute for life style changes in those patients for whom these would be the most appropriate form of intervention? Are we not contributing to a culture of medicalization when spending more quality time with out patients and helping them face the roots of their problems might well be more helpful to them?

We have been brave enough to break the barriers of “protective silence” with our seriously ill patients. We should continue on a path of truth and truth telling in the patient-doctor relationship. This is the foundation for reciprocal trust and a therapeutic relationship. There is a therapeutic effect inherent in the patient-doctor relationship. It is not a placebo effect, but the therapeutic power of human connectedness, compassion and caring.


Daugherty CK, Ratain MJ, Emanuel EJ. Ethical, Scientific, and Regulatory Perspectives Regarding the Use of Placebos in Cancer Clinical Trials. J Clin Oncol 2008; 26: 1371-1378.

Harris G. Half of doctors routinely prescribe placebos. The New York Times, October 24th, 2008.

Surbone A. Truth telling to the patient. JAMA 1992; 268:1661-1662

Surbone A. Telling truth to the cancer patient. What is the truth? Lancet Oncology, 2006; 7: 944-950.

Tilburt JC, Emanuel EJ, Kaptchuk TJ, Curlin FA, Miller FG. Prescribing placebo treatments: results of a national survey of US internists and rheumatologists. BMJ, online publication October 23, 2008.

Wagstaff A. When is it OK to randomize cancer patients to placebo? Cancer World 2008; 27: 20-224.

2 comments on “Is prescribing placebos an ethical practice?

Comments are closed.