The Side Effects of Truth

WE LIVE IN THE pharmaceutical era of medicine, a time of tablet-sized miracles and blockbuster serums. More than 70 new drugs are approved every year, adding to the thousands for which American doctors already write some 3 billion annual prescriptions. The medications prolong countless lives and cause millions of harmful side effects. For most patients, the benefits far outweigh the dangers. An aging man will risk diarrhea to restore his virility. A cancer patient will lose her hair in the hope that chemotherapy will save her life. FDA safety officers like Graham spend their lives searching out the other type of pills, the unexpected killers that harm patients after the FDA has approved them.

Since 1988, Graham has called for the removal of 12 drugs from pharmacy shelves, leading to 10 recalls that have likely saved hundreds, if not thousands, of lives. Each recall is an embarrassment for Graham's employer, the FDA's Center for Drug Evaluation and Research, which approved the drugs in the first place. The trouble is that the roughly 2,300 staffers who support the approval process, and the 109, like Graham, who study the safety of drugs after their release, all fall under the same leadership, and that leadership is highly responsive to industry. In recent years, nearly half of the center's $400 million budget has been paid for by drug companies. This arrangement stems from a 1992 agreement, made partly at the urging of AIDS activists, that the FDA would speed up approvals in exchange for "user fees" from industry. "The focus at FDA is efficacy," says Dr. Curt Furberg, a scientist at Wake Forest University who advises the agency. "Safety is a stepchild."

For the pharmaceutical companies the system works just fine. "The drug-approval process [in the United States] is second to none," says Jeff Trewhitt, a spokesman for PHRMA, the drug industry's trade group. "This is an exhaustive process."

Companies eventually recall about 3 percent of their drugs for safety reasons. Behind many of these recalls are scientists like Graham, who often find themselves pitted against their own supervisors. "If you say something negative about a drug, they try to shut you up," says Dr. Sidney Wolfe, an FDA watchdog for the group Public Citizen, which has been exposing dangerous drugs for three decades. "David is not the only one, by any means, who has raised issues that later proved to be correct."

In 1998, for instance, an FDA drug reviewer named Dr. Robert Misbin wrote a paper showing that the diabetes drug Rezulin had caused liver failure in a patient during a controlled study. When his bosses tried to prevent him from publishing, Misbin saw firsthand how the system encouraged the sacrifice of public health to the interests of the industry. "One of my supervisors said something to me that I have never forgotten," Misbin says, "that we have to maintain good relations with the drug companies because they are our customers." Misbin eventually went public with his concerns, and the drug was pulled a year later. But he has paid for sticking to his principles. "I am no longer given any good projects," he says.

Dr. Andrew Mosholder, another FDA reviewer, faced similar pressures last year when he completed a study showing that antidepressants increased suicidal behavior in children. Further studies proved that Mosholder's science was spot on. But his bosses told him not to report the findings. When someone with access to the study passed his results to the press, the FDA launched an investigation into the leak. According to Tom Devine at the Government Accountability Project, who later became Graham's lawyer, several scientists were interrogated and threatened with possible jail time.

Such intimidation has worked. In 2002, about one in five FDA scientists told federal investigators that they felt pressure to approve drugs despite reservations about safety and efficacy. Two-thirds said they lacked confidence that the agency adequately monitors drug safety after approval.

DAVID GRAHAM IS A CHILD of the Bronx. One of six children, he spent his teenage years in a cramped home in northern New Jersey, sleeping three to a room. It wasn't until he reached college, at Franklin & Marshall, that he decided to become a doctor. Johns Hopkins Medical School offered him a chance at early admission, and he accepted. His plan was to become a general practitioner in Vermont, where he'd live with Nancy Peterson, a girl he had met in the freshman dormitory. He graduated from med school with the second-highest grades in his class–not the highest, he points out, owing to a single B in the medical history course he cut every other Friday, when he'd ride a bus 75 miles to spend the weekend with Nancy. She married him a year later. "It was worth it," he says.

Hopkins, however, was geared toward academic research, not producing family doctors for the Green Mountain State. And Graham never took to working with patients anyway. "What I really enjoyed was problem solving," he says. So after residencies in internal medicine at Yale and neurology at the University of Pennsylvania, Graham took a job at the FDA in epidemiology. His first major safety review, in 1988, focused on a pill called Accutane, intended for patients with severe, previously untreatable acne. The study proved to be a crash course in the ways of the FDA.

At the time, Accutane was seen as a miracle cure by people just like Graham, who still bears the scars of teenage acne. But the side effects could be horrible. One in four children born to women who got pregnant while using the medication suffered from birth defects. Many women didn't bring their babies to term at all, losing them to miscarriage, and most pregnancies ended in abortion. Graham looked at the data and sounded an alarm. Attempts to educate young women about Accutane's risks were failing. "Only the immediate withdrawal of Accutane from the market will work," he wrote to his supervisors in 1990. "The delay only compounds the body count."

That statement was both scientific and political. Graham is opposed to abortion, and the bodies he was referring to were those of the thousands of unborn children whose mothers had taken the drug. Devoutly Catholic, Graham keeps a postcard image of Jesus on his office wall and speaks easily about his calling to service as a Christian. He taught himself ancient Greek so he could read the New Testament as it was written. For him, the scientific process is an extension of his faith. "He really does believe that to know what is real is to know something about the Author of the truth," says John Cavanaugh-O'Keefe, a pro-life activist who met Graham at a prayer group.

But for his managers at the FDA, under pressure from dermatologists and Roche, the drug's manufacturer, Graham's religion was an irrational excuse for his unreasonable claims. "They'd call him the right-to-life nut," says Rep. Bart Stupak (D-Mich.), who recently introduced legislation to restrict patient access to Accutane. "But everything he said has pretty much been proven."

The FDA sided with Roche, rejecting Graham's calls to withdraw the drug and his later pleas to restrict its distribution. In the years that followed, the drug's use among young women nearly tripled, and Roche's annual Accutane revenues exceeded $1 billion. The company designed a voluntary program to encourage birth control among users, but the number of pregnancies continued to rise. Nonetheless, Roche described the program as a success, noting that occasional pregnancies were the inevitable result of birth-control failures. "You can't control everything when you are dealing with human behavior," explains Gail Safian, a Roche spokeswoman.

In 2002, when Roche's patent expired, the FDA helped create a stricter program to ensure that women on the drug used birth control, requiring monthly proof of a negative pregnancy test. Two years later, a panel of scientists found the system was still failing. In November, as the FDA was attacking Graham for his work on Vioxx, the agency announced still stronger restrictions. These measures echo changes Graham asked for more than a decade ago.