If your doctor has prescribed antibiotics for the flu or told you to try B12 vitamins for fatigue, those treatments were probably just a placebo - an unproven therapy offered with the hope you would feel better if you took something.

Treatment with placebos is far more common than you might think, according to a new national survey in which 46 percent to 58 percent of U.S. physicians admitted using placebos regularly. Only 5 percent said they tell patients explicitly that they are doing so.

"Frankly, I was astonished," said Dr. Ezekiel Emanuel, director of the National Institutes of Health department of bioethics and a co-author of the report published Friday by the British Medical Journal. The survey was sent to 1,200 internists and rheumatologists; 679 responded.

Placebos harness the mysterious power of the mind to heal the body; their benefits come from the positive expectations of the patient rather than any physical mechanism. The classic examples are sugar pills, saline injections and other treatments that seem to inspire confidence even though they are inert.

But the physicians surveyed were far more likely to use active agents as placebos, including over-the-counter painkillers, vitamins, sedatives and antibiotics.

What classifies them as placebos is the context. If the recommended treatment hasn't been shown, physiologically, to work for the condition in question, that's a placebo.

While apparently common, treating patients this way is controversial. Two years ago, the American Medical Association said it was wrong to use placebos without a person's knowledge. Some experts consider placebos unethical because they can involve a measure of deceit.

But other experts make a case for what they call "benevolent deception" - letting a patient believe she's getting a useful treatment because, paradoxically, it might work.

Say a person with a viral upper respiratory infection walks into a doctor's office and demands antibiotics even though there's no evidence that these medications, which fight bacterial infections, do anything to help viral illnesses.

"Most of the time, I'll tell patients, 'You really don't need an antibiotic.' But if someone's really nervous or distressed, I'll write a prescription, asking them to wait a day or two and then take it if they don't improve," said Dr. John Hickner, a professor of family medicine at the University of Chicago.

"I guess I do it because if they think it will help them it may, indeed, have a placebo effect," said Hickner, who co-wrote a report on Chicago doctors' use of placebos that appeared this year in the Journal of General Internal Medicine.

Several studies have found that patients respond positively to placebos upward of 30 percent of the time, though others conclude the effects are minimal. Meanwhile, many patients suffer from conditions for which there are no scientifically proven treatments, including respiratory disorders, chronic pain and gastrointestinal distress.

"Doing anything at all tends to help people feel better," said Dr. Farr Carlin, assistant professor of medicine at the University of Chicago and a co-author of the new study.

Physicians often say something like, "I don't really have faith that this will make a lot of difference, but go ahead and try it and see if it makes you feel better," said Dr. Eric Ruderman, an associate professor of medicine who specializes in rheumatology at Northwestern University's Feinberg School of Medicine.

That's a way of walking the line between "I don't think this will help you" - a statement that could undercut the placebo effect - and "I'm pretty sure this will help," which wouldn't be true.

In the new survey, more than two-thirds of physicians said they would tell patients the substance was "medicine not typically used for your condition but [that] might benefit" the patient. Eighteen percent said they'd just call it medicine and 9 percent opted for "medicine with no known effects for your condition."

Dr. Matthew Johnson, a Park Forest family physician, said he would never give someone a sugar pill or lie to a patient, telling them that he believed a treatment would be effective when it wasn't true. "The patient has a right to know if you're giving them something that has some potential for efficacy," he said.

Johnson has, however, given vitamin B12 to patients with chronic fatigue who think the supplement will help and who also agree to improve their diets and exercise more. Similarly, he has agreed to patients' requests to take garlic extract for high cholesterol - along with diet, exercise and perhaps a prescription medication.

"I don't believe the garlic will physically help the patient, but it may motivate them to comply with the rest of my recommended regimen, and that's a good thing," the doctor said