Some 97 percent of British family doctors admit giving patients placebos, and more than three-quarters say they prescribe them at least once a week, a survey released on Thursday found.
Researchers at the English universities of Oxford and Southampton found that the vast majority of general practitioners (GPs) had used "impure" placebos, or unproven treatments.
These include low doses of effective medicines, antibiotics for suspected viral infections, or nutritional supplements not shown to help a particular condition.
Some 12 percent of doctors had used "pure" placebos, such as sugar pills or saline solutions, which contain no active ingredients at all, according to responses from a representative sample of 783 doctors surveyed online.
Administering placebos in general practice contradicts ethical guidelines set out by the General Medical Council, Britain's regulator for doctors.
Placebos were given to "induce psychological treatment effects" because patients pushed for treatment or to reassure them, the doctors told researchers, while the study added that levels were similar to that found in other countries.
But more than 90 percent of doctors objected to placebo use where it could endanger trust between patient and doctor, while more than 80 percent said they opposed it where deception was involved.
"This is not about doctors deceiving patients," said Jeremy Howick, co-lead author of the study.
"The study shows that placebo use is widespread in the UK, and doctors clearly believe that placebos can help patients."
Placebo use outside clinical trials -- where they are used for comparison purposes to test the effectiveness of a treatment, and where patients give informed consent -- is generally considered unethical.
A 2008 US survey focusing on rheumatologists and internal medicine practitioners found that about half used placebos, in contravention of guidelines.
The British Medical Association, the professional body for doctors, said it opposed such use.
"The BMA believes there are fundamental problems associated with doctors prescribing placebos. And in our view the unacknowledged use of placebos for patients with capacity is unethical," a spokesman said.
"Prescribing and administering a placebo must entail some degree of patient deception because to maximise the placebo effect, a patient needs to believe that the 'dummy' treatment administered is real.
"Deceiving a patient, even where the doctor is acting in his or her best interests, obviously undermines trust and risks damaging the doctor-patient relationship."
But Howick, of Oxford University, said ethical guidelines should be revisted in light of "strong evidence suggesting that doctors broadly support their use".
Some 66 percent of respondents to the survey said that pure placebos were ethically acceptable under some circumstances, but 33 percent said they were never acceptable.
Impure placebos, which can also include unnecessary physical examinations or blood tests, were deemed acceptable by 84 percent of doctors, according to the research published in open access journal PLOS One.