What are the risks?
"When you look at the reviews, it appears that about six out of 10 patients do achieve some level of benefit. But the issue is, how many of those six are achieving a benefit that they wouldn't achieve with a placebo or some other approach?" asks Roger Greenberg, PhD, a psychologist at SUNY Upstate Medical University at Syracuse and author of From Placebo to Panacea. "I'd say that out of the six, only about two are receiving a unique benefit from antidepressants, and even that is questionable."
Should people who feel better on Prozac shred their prescriptions? Not just yet. This debate is far from settled.
Some leading psychiatrists and mental health advocates are outraged at the views of Glenmullen and other critics. "Patients say to me, 'I feel normal for the first time in my life,' " says Harvey Ruben, MD, a clinical professor of psychiatry at the Yale School of Medicine.
Ruben calls much of Glenmullen's book "pure speculation." Yes, the antidepressants cause side effects in some patients -- every drug does. "For any drug, if you look in the Physician's Desk Reference, there are probably 150 side effects, many of them so severe that you'd probably never take the drug if you read them," he says. "Glenmullen has taken legitimate side effects, which are very rare, and some terrible case histories, and written a book that makes the uninitiated reader believe that these side effects happen to everybody."
Several psychiatrists also dispute the studies that question the clinical effectiveness of antidepressants. "If you look at all of the studies in the world literature, there are no studies in which a placebo is better than an antidepressant," says Columbia psychiatrist Frederic Quitkin, MD. "In 60% to 70% of the studies, the antidepressant performs better than a placebo. It's simply impossible for that to happen by accident."
What about the studies that seem to indicate that placebos work just as well as antidepressants? Quitkin believes that some of these results may be attributable to flaws in the way the studies were designed and conducted. If a study isn't long enough, or uses an inappropriate dose, it may show a drug is less effective than it really is.