Two years ago, I traveled to Boston and consulted a top-notch academic surgeon over a small, chronic problem that had worsened and was threatening my overall health. The doctor decided to temporize by prescribing a medication he had studied in the course of his research. The drug was not formally indicated for my ailment, but he thought it might help.

I asked, What side effects should I expect?

He answered: None.

Respectful as I was of this surgeon's expertise, I must have shown some sign of doubt.

He repeated: We've given this medicine to hundreds of people. It has no side effects.

Now, this substance was one that re-regulates neurotransmitters. The formulation was supposed to be quite specific - to work its effects locally, in one target organ, avoiding the brain and the rest of the body; but I knew that specificity only ever goes so far. Still, I was content to take the sample packet and walk out the door.

Later that day, I expressed jealousy to a psychiatric colleague. I said, I wish I worked in a specialty where the doctors get to say: This drug has no side effects.

Of course, the reference books offered contradicting testimony. The drug had a number of side effects. Taking the pill, I got plenty, including ones not on the list.

All the same, that surgical boldness was useful to me. I took the message to be: We're all big boys here. I'm the expert. I think you should take this medication. It's reasonably benign. And if you get some headache or nausea or whatever, I want you to tough it out. We're applying a tolerable remedy to a serious problem.

A year later, I saw the surgeon again. By then, he had devised an alternative approach to my ailment. He applied it. I got better. And I got to come off the side-effect-free drug. I felt worlds better. I had, in retrospect, suffered even more adverse effects than I had been aware of — or than I had associated with the pill. All the same, that soldierly, know-nothing reassurance - no worries, mon - had got me through the year and done me good.

How badly had I been fooled? We can all Google a drug and find the standard warnings. The surgeon had let me be of two minds, aware of the risks and inattentive to the nuisance.

Two publications brought that initial office visit in Boston back to mind. The first is Judith Warner's recent blog entry for the New York Times, "Adverse Effects." Warner, whose perspective and writing I admire, tells of an encounter with a neurologist who denies that the medications he prescribes (for migraine) cause side effects. But on the new drug, Warner finds that her brain has fogged over. She can't write, she can't function in her family. She's read the full list of side effects, so she tapers off the medication. Her brief is for doctors to be more open about what can go wrong.

That's how I practice. I try to pass on substantial information, within the limits of what I think a patient can absorb. As treatment continues, I expand the discussion and mention yet more side effects. . . which is all well and good, except that you lose some of the magic. I'm being collegial with patients, showing them due respect, bringing them into the decision-making process. I'm avoiding the doctorly hubris that harmed Judith Warner. But there's a loss here as well. We miss out on some of the placebo effect - on the similar doctorly hubris that worked for me as a patient.

Today, the BMJ, the former British Medical Journal, published results of a survey of American internists and rheumatologists; most said that they prescribe placebos, usually vitamin pills or analgesics. Most thought that the practice was ethically permissible and potentially effective. Impicitly, they were making an old-fashioned case for paternalism and benign deception: these expedients help patients. One specific example was using a sugar pill for fibromyalgia. In terms of what a doctors actually say, according to the study they might describe a vitamin as "a potentially beneficial medicine not typically used for [your] condition."

The survey effectively replicates a study with similar results conducted in Israel and published in the BMJ four years ago. There, most specialists, GPs, and nurses had prescribed placebos, and most practitioners thought that they worked. The indications were, "after ‘unjustified' demand for medication," to calm a patient, to relieve pain, or as a diagnostic tool. Research has found similar outcomes in the United Kingdom, New Zealand, and Sweden. And almost all Danish doctors (86%) seem to recommend placebos.

In the good old days, when psychoanalysis was all-powerful, we psychiatrists had the mojo. Now it's the GPs and joint doctors. Yes, yes, to be sure, our patients are the better for our honesty and humility. But the rheumatologists ,surgeons, and neurologists must know something, too. To say that a drug has no downside, or to give one that has no direct physiological action - as I say, it all makes an office psychiatrist envious . . . and not at all certain that his own practice, for all its seeming rectitude, is as effective as it might be.

Postscript: Readers are free to send notes saying that the drugs psychiatrists prescribe (or the psychotherapies we offer) are all placebos anyway. I've tried to address that contention in this blog. But what's at issue in the BMJ studies is a different ethical issue, the one arising from doctors' recommending a treatment that they themselves believe acts only through creating positive expectancy.

Most Recent Posts from In Practice

A Tribute to Tom Magliozzi

An appreciation of Click and Clack

Bell of Freedom

Pete Seeger

Kenneth C. Edelin

Remembering a Women's Rights Pioneer