Depression Drugs to Carry a Warning

Because the available studies on the drugs were not specifically designed to examine suicide risks, evaluating the scientific evidence has been complicated. Prozac has been shown to be effective in treating children's depression, so today's warning suggests the drug might increase suicidal thoughts and behavior among some children and adolescents while simultaneously relieving depression.

None of the other antidepressants has been specifically approved to treat depression among children, so doctors who prescribe them are extrapolating from studies that show they are effective in adults. In children with depression, the overwhelming majority of clinical trials have failed to show that widely prescribed drugs including Paxil, Zoloft and Effexor are superior to placebos; until recently, however, physicians did not have access to most of those negative results.

"To give a child a drug that can make them suicidal for conditions where there is nonexistent or inadequate proof that the drugs work is irresponsible and fails to protect our children," said Joseph Glenmullen, a Harvard University psychiatrist and the author of "Prozac Backlash," who urged the FDA to go further in limiting the use of unapproved antidepressants.

FDA officials said patients are to receive a separate warning, called a "MedGuide," written in easy-to-understand language, about the risks.

The changes will be implemented immediately, officials said, and patients and doctors should see the new warnings within weeks. All advertising for the drugs will have to mention the risks of suicidal thinking and behavior.

Physicians are not likely to be told exactly how many studies of a given drug turned out negative. The problem with describing positive and negative studies, said Robert Temple, director of the FDA's Office of Medical Policy, is that doctors may think a medication is effective because of one positive study. But the FDA requires two positive studies to approve a medication.

Kweder said the wording of the labeling will be worked out in consultation with individual companies, which have sometimes evaluated their products very differently than the agency. In the case of Zoloft, for example, Pfizer Inc. published a paper that pooled the results of two trials and concluded that the results were positive. FDA regulators, however, ruled that individually, the two trials were negative.

"It'll be our judgment" as to what goes on the label, Temple said yesterday.

Regulators also said they were examining data from trials of the drugs in adults to see if there was a similar pattern of suicidal tendencies.