Depression pills: Do they kill or cure?

Only hours before Stephanie Fritz hanged herself last November, she was sitting at the dinner table with her family in Tinton Falls, excitedly discussing an upcoming vacation and a concert at the Stone Pony in Asbury Park. The 15-year-old high school sophomore, a member of the swim and tennis teams at Red Bank Catholic, had been taking Zoloft, a widely used antidepressant, for four months. On this day, her last, she seemed upbeat.

Stephanie's father, Scott, blames the medication for his daughter's fatal mood swing. "The drug companies say kids get depressed and that suicide is a byproduct of the illness," he said. "But no one knows how these drugs work. And I think there's a definite connection."

Fritz's suspicions are at the heart of a growing controversy over a popular group of medicines that also includes Prozac, Effexor and Paxil. Researchers, who do not fully understand the effect of antidepressants on the brain, have thus far failed to prove a link between the drugs and suicidal behavior.

Going forward, the effort to establish any such link may be hampered as much by ethical dilemmas as by scientific quandaries. Researchers have access to data on suicide (30,622 deaths in the United States in 2001, with 1,611 victims between ages 15 and 19), and to rough estimates of antidepressant use (up 9.2 percent each year from 1998 to 2002, according to the pharmacy benefit manager Express Scripts Inc.). But, ethically speaking, should anyone run a study designed specifically to determine if a drug might kill a kid?

"I'm not sure these questions really can be answered. We can only look at associations," said Aidan Stokes, a psychiatrist and member of the Canadian government advisory panel that has warned against prescribing the drugs to children and adults. "But I think the apprehension has to be addressed."

The issue has engulfed regulators, doctors and lawmakers, and has turned a light on the often-ignored world of unpublished clinical trials. Drug companies argue strenuously that their trials -- published or unpublished -- have not produced any suicides. But last week, the American Medical Association called for the creation of a public registry for all trials, out of concern that negative data is suppressed by drug makers.

"Studies with positive findings are more likely to be published than studies with negative or null results," says Joseph Heyman, an AMA trustee. "We are concerned that this pattern of publication distorts the medical literature."

The debate over antidepressants and suicidal behavior, which surfaced a decade ago after Prozac was approved by regulators, resurfaced last year when medical authorities in Britain issued warnings about prescribing the drugs, especially to anyone under 18. In the United States, only Prozac is approved for use in patients under 18, but other antidepressants are commonly prescribed -- and some experts have voiced concern that children are especially susceptible to the effects of the drugs.

Under growing pressure, the Food and Drug Administration held a meeting earlier this year to explore the issue. Fritz and many other parents whose children attempted or committed suicide while on an antidepressant testified that the drugs should be restricted or banned.

The agency, which is being investigated by a congressional committee for allegedly muzzling one of its own medical reviewers at that meeting, did not reach a conclusion. Two months later, however, the FDA did tell drug makers to strengthen warnings on product labels.

Meanwhile, the FDA commissioned Columbia University researchers to review reports on side effects and clinical trial data from drug companies. The results are due this summer, but there are no guarantees this undertaking will yield anything conclusive. And even if it does find links to suicide, the FDA appears unlikely to withdraw any of the antidepressants.

"I don't want to rule out any possibility, but it's very difficult to imagine these drugs would be withdrawn. They've all been very effective in adults," said Russell Katz, who heads the FDA's division of neuropharmacologic drug products. "But is it possible we won't have a definitive decision? Yes, that's absolutely possible."

One reason is that each drug company's clinical trials describe suicidal behavior differently. A key challenge for the Columbia researchers is to scour the data and distinguish between patients who made a serious attempt at suicide and patients who harmed themselves only minimally.

Given the difficulties of this analysis, "it's more likely that they can demonstrate an association" between antidepressants and suicide, but not a definitive scientific link, said David Fassler, a psychiatry professor at the University of Vermont and a trustee of the American Psychiatric Association.

To some experts, an association may be enough of a red light.

"Causal links are rarely possible with 100 percent certainty," said Jane Garland, a psychiatry professor at the University of British Columbia in Vancouver and a member of the Canadian advisory panel. "But patterns of association can be seen," she said, and these patterns can "build a reasonably strong case."

As an example, she cited studies conducted by GlaxoSmithKline, which sells Paxil, that showed some mildly depressed adolescents needed to be hospitalized for suicidal ideas or worsening depression after receiving the medicine. But no one taking a placebo was hospitalized.

"That is hard to dismiss," she said. "If these drugs are treating depression, you would expect the opposite -- a lower rate, not a higher rate, in those receiving the drug."

David Healy, a professor at the University of Wales and author of "Let Them Eat Prozac," noted that as more clinical trial data becomes available -- Glaxo last week published summaries on it Web site, for instance -- regulators and physicians will gain new insights.

The drug makers have studies "where they had made what are still confidential judgments that their drugs caused agitation/ suicidality in people," said Healy, who has been critical of the pharmaceutical industry.

Even if the Columbia University analysis suggests a link between antidepressants and suicide, Canada's Stokes believes that further study will be needed. But, he added, it's unlikely that the pharmaceutical industry would be willing to conduct the research.

"It's unlikely that there will be a smoking gun," said Steven Simring, a psychiatry professor at the University of Medicine & Dentistry in New Jersey. "It's really a statistical story. And the best they can come up with is a strong link. But I don't think it will be the final word anyway, because no matter what they say, someone will be unhappy."

Ed Silverman covers the pharmaceutical industry for The Star-Ledger.