Prescriptions for what researchers deemed “essentially placebo,” are given frequently by general practitioners (GPs) in Israel and 20 European countries to a small percentage of patients, according to data from a survey study. Studies in the US have also tracked the use of placebo treatments. 

“This suggests that essentially placebo prescriptions were common at a population level, which poses risks for the patient-GP relationship and creates medical risks for patients,” wrote the authors, led by Fabian Wolters, MSc, with Unit Health, Medical, and Neuropsychology at Leiden University in Leiden, the Netherlands. Findings were published online in JAMA Network Open.

The survey study included 952 respondents across the 21 countries. The median prescription rate for placebo treatments was once every 2 weeks or in 0.7% of all consultations; 84% indicated they had prescribed an essentially placebo treatment at least once.

Definition of ‘Essentially Placebo’

The researchers defined essentially placebo treatments to participants as “when, in your estimation, any positive treatment effect on the patient’s symptoms is not caused by the pharmacological or biological components of the treatment.”

That definition is a key strength of the research, Michael Bernstein, PhD, director of the Medical Expectations Lab in the Department of Radiology at Warren Alpert Medical School, Brown University, Providence, Rhode Island, told Medscape Medical News.

Defining “placebo” has been tricky in past research, he said. The authors wrote that previous studies have drawn a line between “pure” placebos (those with no active ingredients, such as a sugar pill) and “impure” placebos (they have an active ingredient, but it will not help resolve current concerns), such as an antibiotic for a viral infection.

He also praised the gathering of results from 21 countries because previous studies “have been done in one country, which makes comparisons more difficult because methodology differs from one country to the next,” he said.

Bernstein said that having worked in this area, he is not surprised by the frequency of placebo prescribing but said patients might be.

He pointed to some previous papers, among them a 2008 paper in the BMJ that found about half of the 679 surveyed internists and rheumatologists in the US reported prescribing placebo treatments regularly (46%-58%, depending on how the question was phrased).

When to Prescribe Placebo

The American Medical Association (AMA) in its code of medical ethics has guidance on prescribing placebo and said physicians may use placebos for diagnosis or treatment only if they: enlist the patient’s cooperation, obtain the patient’s general consent to administer placebo, and avoid giving a placebo merely to mollify a difficult patient.

“I would say by that standard, my reading of this is that what’s being done probably violates this AMA code of ethics in a great many cases,” Bernstein said. “I think there’s a clear understanding and expectation that an inactive placebo is not ethical. Even an active placebo would not be ethical aside from a very narrow set of circumstances, and I think physicians very rarely meet that very narrow set of circumstances.”

Bernstein said it’s hard to determine the prevalence of prescribing placebo in the US with small numbers of studies across a number of specialties and the definition of frequent placebo prescribing “far from clear.”

He added that there are many cases in which placebo is effective, which complicates the equation. He noted physicians have multiple responsibilities to their patients. One is to give patients treatments to improve their health and that could include placebo. Another duty we accept, he said, “at least in the US, is that physicians also have a duty to be transparent to their patients. There’s a tension between these two duties.” 

Placebo Decision Motives and Outcomes Need Study

In an accompanying editorial,Pekka Louhiala, DMedSci, PhD, with the Faculty of Social Sciences at Tampere University in Tampere, Finland, said a limitation of Wolters’ study is the relatively low number of participants per country and different response rates between countries.

She said future research should focus not on the precise prevalence of prescribing essentially placebo treatments because these findings are clear that the practice occurs regularly. Instead, she wrote, future research should focus on the reasons the decisions are made and the outcomes of the decisions.

“Such research could offer valuable insights into how physicians conceptualize medicine — as both a science and a practice,” she wrote. “The potential impact of these prescriptions on the physician-patient relationship and on trust in healthcare more broadly should also be examined.”

Wolters and Louhiala reported no relevant conflicts of interest. This study was supported by a Netherlands Organisation Scientific Research — Vici grant. Bernstein reported being a co-author of the book “The Nocebo Effect: When Words Make You Sick.”

Marcia Frellick is an independent, Chicago-based journalist and a regular contributor to Medscape.