In June, after the Independent article, the FDA (without issuing a press release) noted on its Web site that one suicide "was reported in a Cymbalta clinical pharmacology study in a healthy female volunteer." The agency added that new data from stress urinary incontinence trials showed that middle-aged women taking duloxetine had a suicide attempt rate of 400 per 100,000 person-years, more than double the rate of about 160 per 100,000 person-years among other women of a similar age. These findings had been withheld from the public, and the researchers asking for them, for five months after the FDA had reviewed data showing the increased risk.
The FDA claims it has no choice but to resist releasing information about drugs it doesn't approve. "My hands are tied," said Dr. Robert Temple, FDA's director of medical policy. "This is something only Congress can change." That may be as much a matter of the FDA's interpretation as it is of the law, however. Experts disagree about whether congressional action or a federal court ruling is needed to make data like Johnson's death available, or whether the FDA could choose to disclose more itself.
The voluntary guidelines promoted by the drug industry, however, are not a solution. These guidelines encourage companies to list every clinical trial they initiate. Registration would be helpful. But it would not compel companies to release the data from, or even the outcomes of, their trials, as long as they companies can argue that this information is "commercially protected."
The use of trade-secret laws to conceal deaths and serious side effects linked to drugs has the obvious flaw of putting profits before public health. It also subverts the covenant between researchers and study volunteers. Subjects like Traci Johnson are told that even if they do not personally benefit from a new drug, the scientific knowledge gained from the study in which they've participated will benefit others. The volunteers should be told instead that scientists will learn about their experience only if it's good news for the drug they're helping to test.
Correction, Nov. 2, 2005: The sentence originally stated that Shionogi & Co. partners with Lilly to market duloxetine. The companies have a licensing agreement rather than a partnership. ( Return to corrected sentence.)