and Leah Driggers-How safe is safe enough? The U.S. Food and Drug Administration (FDA) has released to the American public an array of truly beneficial drugs in recent years, but it has also approved-then quickly "unapproved"-at least 10 drugs that shortly after their market debuts started to kill and maim people who took them. In June 1998, for example, just a year after approving the high blood pressure medication Posicor, the FDA asked the drug's maker to withdraw it because of potentially deadly interactions with 25 other drugs. Rotashield, a vaccine the FDA approved in 1998 for severe childhood diarrhea, was withdrawn in 1999 after causing bowel obstruction. In October 1999, two years after the agency released antibiotic Raxar to consumers, its manufacturer withdrew it after patients taking it died of irregular heartbeats. FDA-approved medications kill more than 100,000 Americans each year, according to a 1998 report published in the Journal of the American Medical Association. That makes lethal drug side effects-not including prescribing errors or drug abuse-the nation's sixth leading cause of death. Spiraling fatalities and an utter lack of public accountability have consumers, drug experts, and even FDA employees questioning the nation's drug-approval system. Longtime FDA pharmacologist Elizabeth Barbehenn, fed up with the pressure to skim over safety concerns, left the agency in 1998. She said the unofficial "message" inside the FDA was that drug reviewers "should be approving things, not questioning problems that arise, and ... give the drug company the benefit of the doubt." FDA reviewer Michael Elashoff left the agency last year. "People are aware that turning [a drug] down is going to cause problems with officials higher up in FDA, maybe more problems than it's worth," he told the Los Angeles Times. Nearly every drug carries with it the risk of harmful side effects. Part of the FDA's job is to weigh those against benefits patients may gain from new medications. FDA reviewers admit it is sometimes difficult to say when risk overrides benefit, especially when there is no other treatment available for a particular illness. But the official measuring stick is more wobbly than American consumers might expect: "No drug is absolutely safe.... There is always some risk of an adverse reaction," states a report from the Center for Drug Evaluation and Research (CDER), the FDA's drug-approval arm. "However when a proposed drug's benefits outweigh known risks, the FDA's CDER considers it safe enough to approve." But how safe is "safe enough?" The agency, long criticized for delaying approval of potentially lifesaving medications while patients died waiting, tried during the 1990s to improve its system. In 1992, the agency worked with Congress and the pharmaceutical industry to shape the Prescription Drug User Fee Act. Legislators designed the act to speed the FDA's historically slow drug review process so doctors could speed potentially life-saving drugs to seriously ill patients. The law gave drug companies opportunity and incentive to subsidize the CDER with "user fees"-money the agency would use to hire more medical officers, and thereby accelerate drug review and approval. For its part, the CDER agreed to review priority drugs (those that are potentially major advances in health care) in six months, and standard drugs (those that offer only minor or no improvement over drugs already on the market) in one year. In 1997, CDER agreed to even faster review times for drugs that target serious and life-threatening illnesses like AIDS. For both CDER and the pharmaceutical industry, the new, hurry-up laws seemed a win-win solution. Drug approvals doubled and CDER cut its per-drug review time in half. But the faster-is-better ethos may be at the root of the FDA's current crop of bad medicine. Since 1996, the agency has had to ask drug makers to yank from the market a slew of medications-sometimes within months of the agency's having approved them-after patients died or suffered serious side effects from taking them. Case in point: Redux, a weight-control aid the FDA signed off on in 1996. Just 15 months after approving it, the agency asked the drug's maker to pull it from circulation because it caused serious heart valve damage in many patients who took it. In February 2000, the FDA approved the irritable bowel syndrome medication Lotronex. After six months on the market and 19 reports of severe intestinal side effects, the CDER ordered manufacturer Glaxo Wellcome, Inc. to attach to every Lotronex bottle a "medication guide" detailing side-effect risks to patients. Meanwhile, pharmacists continued to fill prescriptions-500,000 of them. But nine months after Lotronex first hit the market, Glaxo Wellcome yanked the drug at the FDA's request. During the drug's brief circulation, at least eight women died (Glaxo calls these deaths "rare reports of fatal sequelae") of severe intestinal damage while taking it. The General Accounting Office (GAO) reported last year that CDER wasn't properly tracking drugs for problems after they hit the market. That could be because the mushrooming CDER approval structure-funded by drug company user fees-has outgrown the agency's ability to follow up on products it approves. CDER's "drug policing" wing, called Postmarketing Drug Risk Assessment (PDRA), conducts "postmarket surveillance" by tracking and monitoring drugs after they are released for public consumption. But though CDER boasts nearly 2,000 employees, only 80 work in PDRA. And while "user fees" paid by pharmaceutical companies speed up drug approvals, not a penny of such fees is authorized for use in tracking drugs' postmarket performance. Former FDA commissioner Carol Scheman, who left the agency in 1994, says CDER's postmarket surveillance is "wildly underresourced and underused." Compounding the problem: Doctors and hospitals are not required to report difficulties their patients encounter with specific medications. Reporting is purely voluntary, and the FDA is aware of the disconnect. Last year, the agency proposed establishing "sentinel" hospitals, medical facilities specially tasked with tracking the performance of new drugs. President Clinton asked Congress to give the FDA an additional $15.7 million to improve side-effect reporting, including adding the sentinel hospitals, but legislators turned him down. So how does the FDA find out about problem drugs? Through consumers or self-reporting by drug companies. Who reviews such negative reports? Usually, the same CDER team that initially approved the drug. Some FDA critics question the circuitous nature of the agency's quality control system and wonder whether the federal government has proven itself a reliable arbiter of quality at all. "Because the agency is not accountable to market [forces], it does not have the incentive to get products out efficiently as possible, nor incentive to withhold shoddy products, or ensure proper labeling," said Ed Hudgins, director of regulatory studies at the Cato Institute. Take Lotronex, for example. When the FDA first released the drug, Philip Anderson, director of the Drug Information Service at the University of California San Diego, carefully examined Glaxo Wellcome's package insert. Mr. Anderson told WORLD that the literature did indicate "rare" instances of severe intestinal obstruction, "but that information was buried ... not prominently placed, and the language used indicated that [the FDA] wasn't clear on whether those incidents were related to the drug." Within months, the deaths of eight women confirmed that they probably were. Mr. Anderson, who has worked in pharmacy for than two decades, says "there is something to the argument that drugs are being rushed through a little too fast." What can be done to improve the FDA's recent safety record? That depends on whom you ask. Mr. Hudgins believes the double-edged sword of free-market accountability would serve patients and the pharmaceutical industry better than a federal agency ever could. Since pharmaceutical companies and private labs perform safety and efficacy testing, he suggests that private labs test and certify drugs and let consumers-the people whose health and lives are on the line-choose which laboratories to trust. Mr. Hudgins believes private-sector drug approval would promote competition, providing economic incentive for full disclosure of a drug's pros and cons. Doctors would then pass that information on to consumers who, based on a combination of medical information, the doctor's recommendation, and the reputation of the approving lab, would have the final say over which medications to take. But the American Enterprise Institute's John Calfee isn't persuaded that the FDA itself is the root of the problem. "It is very unlikely the FDA is negligent," he told WORLD. "The agency has no incentive to rush drugs to market prematurely." Mr. Calfee believes factors outside the FDA's influence-including "off-label use," in which doctors prescribe drugs to treat conditions for which they were not originally designed-may be to blame for problems with recently approved drugs. "Many of the adverse events we're seeing are not because anything is wrong with the drugs intrinsically, but because they're used improperly. I think there's a lot of good to be gained by close monitoring of how drugs are used in the field." For now, FDA critics shout from opposite sides of a deep ravine. One camp blasts the agency's newer, rapid release of sometimes-fatal drugs that treat nonfatal conditions. The other camp, which includes patient advocacy groups, still accuses the FDA of dragging its feet on approving lifesaving medications. It's the pharmaceutical version of a military conundrum: Should the FDA both hurry up and wait?