Skip to main content

Are Antidepressants Just Placebos with Side Effects?

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


I have first-hand experience of the devastation of depression, in myself and those close to me. Although I have been tempted to try antidepressants, I've never done so. Of course, like everyone reading this column, I know many people who have been treated with antidepressants—not surprisingly, because according to a 2005 survey, one in 10 Americans are now under such treatment. Some people I know have greatly benefited from their treatment. Others never find adequate relief, or they experience annoying side effects—such as mania, insomnia, emotional flatness or loss of libido—so they keep trying different drugs, often in combination with psychotherapy. One chronically depressed friend has tried, unsuccessfully, to stop taking his medications, but he experienced a surge of depression worse than the one that led him to seek treatment. He accepts that he will probably need to take antidepressants for the rest of his life.

We all, to greater or lesser degrees, have this kind of personal perspective on antidepressants. But what does research on these drugs tell us about their efficacy? The long-smoldering debate over this question has flared up again recently, with two medical heavyweights staking out opposite positions. In a New York Times essay, "In Defense of Antidepressants," Peter Kramer, a professor of psychiatry at Brown, insists that antidepressants "work—ordinarily well, on a par with other medications doctors prescribe."


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Kramer's article seeks to rebut a wave of negative coverage of antidepressants, most notably a two-part essay in The New York Review of Books (which can be found here and here) by Marcia Angell, former editor of The New England Journal of Medicine and now a lecturer in social medicine at Harvard. Angell cites research suggesting that antidepressants—including both selective serotonin reuptake inhibitors (SSRIs) and other medications—may not be any more effective than placebos for treating most forms of depression.

Angell highlights a meta-analysis, carried out by the psychologist Irving Kirsch, of trials of a half dozen popular antidepressants submitted by drug companies to the U.S. Food and Drug Administration. Many of the studies were never published because they failed to yield positive results. (The practice of burying negative results from trials is still quite common, as this recent Scientific American blog post points out.) After analyzing all the FDA studies, Kirsch concluded that placebos are 82 percent as effective as antidepressants. According to Kirsch, this difference vanishes if antidepressants are compared to "active placebos," which are compounds such as atropine, an alkaloid that blocks certain nerve receptors and causes dry mouth and other symptoms, that have distinct side effects.

Angell quotes from Kirsch's new book The Emperor's New Drugs (Basic Books), in which he states that "the relatively small difference between drugs and placebos might not be a real drug effect at all. Instead, it might be an enhanced placebo effect." This "startling" claim, Angell adds, "flies in the face of widely accepted medical opinion, but Kirsch reaches it in a careful, logical way. Psychiatrists who use antidepressants—and that's most of them—and patients who take them might insist that they know from clinical experience that the drugs work. But anecdotes are known to be a treacherous way to evaluate medical treatments."

So how does Kramer begin his defense of antidepressants? With an anecdote—about a friend who benefited from antidepressants after suffering from a stroke. This rhetorical strategy should not be surprising, since Kramer's 1993 bestseller Listening to Prozac (Penguin), which contributed to the surge in popularity of Prozac and other SSRIs, relied heavily on anecdotal evidence rather than clinical data. Kramer told story after story of patients transformed by Prozac. He suggested that SSRIs might be ushering in an era of "cosmetic psychopharmacology" in which patients are not only cured of disorders but become "better than well."

The Brave New World envisioned by Kramer was always a complete fantasy. When he wrote his book in the early 1990s, studies by Eli Lilly, Prozac's manufacturer, showed that it was no more effective than older antidepressants, such as tricyclic drugs, or psychotherapy without drugs. Although Prozac was touted for its relatively mild side effects, it causes sexual dysfunction in as many as three out of four consumers. Kramer relegated a discussion of Prozac's sexual side effects to the fine print, literally, in his book's endnotes. His Times essay doesn't provide any better data for antidepressants than Listening to Prozac did. Kramer delves into an arcane discussion of how difficult it is to distinguish genuine drug benefits from placebo effects, but he does not really grapple with the claim of Angell and Kirsch that antidepressants may be active placebos.

Kramer does not mention, for example, a recent analysis of STAR*D (Sequenced Treatment Alternatives to Relieve Depression), which has been called "the largest antidepressant effectiveness trial ever conducted." According to a group of four researchers, STAR*D data show that "antidepressants are only marginally efficacious compared to placebos," and even this modest benefit might be inflated by "profound publication bias." The authors recommend "a reappraisal of the current recommended standard of care of depression."

Angell agrees. She thinks that the surge in antidepressant prescriptions over the past two decades stems less from the drugs' efficacy than from the marketing muscle of the pharmaceutical industry, which she says "influences psychiatrists to prescribe psychoactive drugs even for categories of patients in whom the drugs have not been found safe and effective."

She recommends that doctors be prohibited from prescribing psychiatric drugs "off-label"—that is, for disorders and populations, notably children and even toddlers, for which they have not been approved. She also urges that we "stop thinking of psychoactive drugs as the best, and often the only, treatment for mental illness or emotional distress.... More research is needed to study alternatives to psychoactive drugs," including exercise and psychotherapy (although of course studies of psychotherapy reveal that it may also work by harnessing the placebo effect).

Given what science is telling us about antidepressants, Angell's recommendations seem wise to me. I sometimes suspect that psychiatric drugs work, to the extent that they do, simply by making people feel different. The suffering person interprets this difference as an improvement, in the same way that someone who is in a rut may feel better by traveling to another country. But does that mean that any psychoactive drug—Caffeine? Beer? Antihistamines? Psilocybin?—can in principle produce the same benefits as an SSRI, as Angell and Kirsch seem to suggest? Even for a skeptic like me, that seems hard to believe. We clearly need more research not only on alternatives to antidepressants (yoga, meditation, jogging, reading groups, journal-writing) but also on the drugs themselves, to understand why some people benefit so much from them while others don't. But more research will be helpful only if the results are reported—as all medical data should be but too often aren't—with absolute candor and transparency.

Photo credit Wikimedia Commons