For years, Johnson & Johnson obscured evidence that its popular Ortho Evra birth control patch delivered much more estrogen than standard birth control pills, potentially increasing the risk of blood clots and strokes, according to internal company documents.
But because the Food and Drug Administration approved the patch, the company is arguing in court that it cannot be sued by women who claim that they were injured by the product - even though its old label inaccurately described the amount of estrogen it released.
This legal argument is called pre-emption. After decades of being dismissed by courts, the tactic appears to be on the verge of success, lawyers for plaintiffs and drug companies say.
The Bush administration has argued strongly in favor of the doctrine, which holds that the FDA is the only agency with enough expertise to regulate drug makers and that its decisions should not be second-guessed by courts. The Supreme Court is to rule on a case next term that could make pre-emption a legal standard for drug cases. The court ruled in February that many suits against the makers of medical devices such as pacemakers are pre-empted.
More than 3,000 women and their families have sued Johnson & Johnson, asserting that users of the Ortho Evra patch suffered heart attacks, strokes and, in 40 cases, death. From 2002 to 2006, the food and drug agency received reports of at least 50 deaths associated with the drug.
Documents and e-mail messages from Johnson & Johnson, made public as part of the lawsuits against the company, show that even before the drug agency approved the product in 2001, the company's own researchers found that the patch delivered far more estrogen each day than low-dose pills. When it reported the results publicly, the company reduced the numbers by 40 percent.
The FDA did not warn the public of the potential risks until November 2005 - six years after the company's own study showed the high estrogen releases. At that point, the product's label was changed, and prescriptions fell 80 percent, to 187,000 by February from 900,000 in March 2004.
Gloria Vanderham, a Johnson & Johnson spokeswoman, said the company acted responsibly. But Janet Abaray, a plaintiff's lawyer from Cincinnati, said that Johnson & Johnson took advantage of an agency overwhelmed by its many responsibilities.
A series of independent assessments have concluded that the agency is poorly organized, scientifically deficient and short of money. In February, its commissioner, Andrew C. von Eschenbach, acknowledged that the agency faces a crisis and might not be "adequate to regulate the food and drugs of the 21st century."
The FDA does not test experimental medicines but relies on drug makers to report the results of their own tests completely and honestly. Even when companies fail to follow agency rules, officials rarely seek to penalize them. "These are scientists, not cops," said David Vladeck, a professor at Georgetown Law School.
Last month, at a trial over the schizophrenia drug Zyprexa, Dr. John Gueriguian, a scientist who worked at the FDA for two decades, testified the agency did not always ask for strong warnings even if it believed a drug was risky. Companies usually oppose warnings, and the agency knows it must compromise on its requests or face years of delay, Gueriguian said. "We at the FDA know what we can obtain and we cannot obtain," he said. "We have many, many problems, and we have a management system - what we can't obtain, we will not ask."
But now, the agency says a proliferation of lawsuits could lead to an overlapping patchwork of rules that would burden companies and might discourage patients from taking useful medicines.
Lawyers for Johnson & Johnson say that patients should not be allowed to sue the company, because the FDA approved the patch and its label.
"FDA is responsible for making those decisions," said John Winter, a lawyer for the company.
Judge David A. Katz of Federal District Court for the Northern District of Ohio is expected to rule on whether lawsuits against Johnson & Johnson can go forward.