Weighing the Risks: The Coming of Age on Antidepressants

“I’ve grown up on medication,” my patient Julie told me recently. “I don’t have a sense of who I really am without it.”

At 31, she had been on one antidepressant or another nearly continuously since she was 14. There was little question that she had very serious depression and had survived several suicide attempts . In fact, she credited the medication with saving her life.

But now she was raising an equally fundamental question: how the drugs might have affected her psychological development and core identity.

It was not an issue I had seriously considered before. Most of my patients, who are adults, developed their psychiatric problems after they had a pretty clear idea of who they were as individuals. During treatment, most of them could tell me whether they were back to their normal baseline.

Julie could certainly remember what depression felt like, but she could not recall feeling well except during her long treatment with antidepressant medications. And since she had not grown up before getting depressed, she could not gauge the hypothetical effects of antidepressants on her emotional and psychological development.

Her experience is far from unique. Since their emergence in the late 1980s, serotonin reuptake inhibitors like Prozac and Zoloft have become some of the most widely prescribed drugs in the world, for depressed teenagers as well as adults. Because depression is often a chronic, recurring illness, there are certain to be many young people, like Julie, who are coming of age on these newer antidepressants.

We know a lot about the course of untreated depression, probably more than we do about very long-term antidepressant use in this population. We know, for example, that depression in young people is a very serious problem; suicide is the third-leading cause of death in adolescents, not to mention the untold suffering and impaired functioning this disease exacts.

By contrast, the risk of antidepressant treatment is small. A 2004 review by the Food and Drug Administration, analyzing clinical trials of the drugs, did show an elevated risk of suicidal thinking and nonlethal suicide attempts in young people taking antidepressants — 3.5 percent, compared with 1.7 percent of those taking a placebo. But since the lifetime risk of actual suicide in depressed people ranges from 2.2 to 12 percent, risk from treatment is dwarfed by the risks of the disease itself.

Still, what do we know about the effects of, say, 15 to 20 years of antidepressant drug treatment that begins in adolescence or childhood? Not enough.

The reason has to do with the way drugs are tested and approved. To get F.D.A. approval, a drug has to beat a placebo in two randomized clinical trials that typically involve a few hundred subjects who are treated for relatively short periods, usually 4 to 12 weeks.

So drugs are approved based on short-term studies for what turns out to be long-term — often lifelong — use in the world of clinical practice. The longest maintenance study to date of one of the newer antidepressants, Effexor, lasted only two years and showed the drug to be superior to a placebo in preventing relapses of depression.

What do I say to a depressed patient who is doing well after five years on such a drug but can’t stop without a depressive relapse and who wants reassurance that the drug has no long-term adverse effects?

I usually say that we have no evidence that the drug poses a risk with long-term use; and since the risk of untreated depression is much greater than the hypothetical risk of the drug, it makes sense to stay on it.

This large gap in our clinical knowledge is compounded by the public’s growing and well-founded skepticism about research sponsored by drug makers. A study in the January 2008 issue of The New England Journal of Medicine, involving 74 clinical trials with 12 antidepressants, found that 97 percent of positive studies were published, versus 12 percent of negative studies.

Clearly, physicians and the public need much better data on the safety and efficacy of drugs after they hit the market, which at present consists mainly of anecdotes and case reports.

Congress recently reauthorized the Prescription Drug User Fee Act, which will expand the F.D.A.’s post-marketing drug surveillance, though I think it did not go far enough in mandating the use of powerful epidemiological strategies to monitor drugs over the long term.

Beyond these concerns, there are other important issues to consider in long-term use of antidepressants, especially in young people. One patient, a woman in her mid-20s, told me that she felt pressured by her boyfriend to have sex more often than she wanted. “I’ve always had a low sex drive,” she said.

For the past eight years she had been taking Zoloft, which like all the antidepressants in its class is known to lower libido and to interfere with sexual performance. She had understandably mistaken the side effect of the drug for her “normal” sexual desire and was shocked when I explained it: “And I thought it was just me!”

This just underscores how tricky it can be to use psychotropic drugs during adolescence — when the brain is still developing, when one’s identity is still work in progress.

The drugs save lives, and we often have no choice but to use them — even if we have questions about their long-term use. But the questions are big ones, and we owe it to our patients to try to answer them.