WASHINGTON -- For years, the Food and Drug Administration has been stymied by medications that are approved for market but later turn out to have serious side effects that its current system of reporting failed to spot.
Now, a proposal that could be a big step toward solving the problem has been offered by the head of the Medicare system.
Whether it's a suspected link between Viagra and rare cases of blindness or heart attacks among arthritis patients using Vioxx, the FDA's current system of voluntary reporting picks up no more than 10 percent of serious drug reactions.
To help shrink the gap, Medicare head Mark McClellan, a doctor and economist who once ran the FDA, has proposed using billing data from Medicare's new prescription drug benefit to track the safety and effectiveness of medications.
McClellan proposes taking billing data from the prescription program, which will begin in January, and combining it with information collected when Medicare users submit claims for hospital and doctor care. By cross-referencing the information, a computer system could spot signs of trouble.
Patients' personal information would be protected, but the data pool would be so large that problems should be much easier to detect, he said.
The problem now is that the FDA approves drugs for market on the basis of trials involving a few thousand patients -- and sometimes dangerous side effects don't show up until a prescription is widely used on the open market.
The Medicare data would cover millions of patients and create an important warning network.
"One of the main frustrations with Vioxx was that it was on the market for a number of years and used by millions of people, and only later was there a possible association with problems," McClellan said.
"There is a legitimate reason for the public to be concerned. In an era when we should be depending on electronic systems, it should not take five years and millions of patients before we get to a potential problem."
The FDA, however, has been noncommittal. "It really is too early," said a spokeswoman, adding: "The agencies are exploring this possibility."
Some outside advisers to the FDA are baffled by what they see as a lack of resolve.
The Medicare idea was one of several discussed at a recent FDA drug safety advisory committee meeting on ways to improve monitoring and reporting.
But after two days of deliberations, "No one at the agency was willing to say, 'OK, this is what we would like to do,'" said Dr. Curt Furberg of Wake Forest University, a member of the advisory panel.
"We thought we had some good ideas, and [using Medicare data] was one of them. It's a little bit frustrating," he said.
The panel's chairman, Dr. Peter Gross of Hackensack University Medical Center, said he thought the Medicare proposal was "a good idea," but "I didn't get the sense that [FDA officials] were latching on to it."
A large private insurer, UnitedHealth Group, is creating a monitoring system for up to 20 million of its patients, similar to the one McClellan has proposed.
The Medicare plan could cover as many as 43 million people, including some of the most vulnerable and sickest that drug companies don't usually enroll in a clinical trial.
The prospect of using Medicare to monitor drug safety has generated interest among medical and pharmaceutical experts.
"It's a huge opportunity," said Dr. Raymond Woosley, vice president of the University of Arizona Health Sciences Center. "We're going to be spending billions of dollars on medications for the elderly, and we're calling it a 'benefit.' But if we don't have a way to monitor outcomes from the use of these medicines, it may not be a benefit."
Woosley, who has testified before Congress on ways to improve drug safety, said he believes the FDA will ultimately endorse the idea.
The centerpiece of the agency's current safety surveillance network is a database called the Adverse Event Reporting System.
Doctors and hospitals voluntarily submit reports, and drug companies are required to pass on those they receive. The system collects about 400,000 reports a year, but most experts say it only scratches the surface.
"We don't know the true extent of underreporting, but we know it's vast," Joyce Weaver, a pharmacologist with the FDA's Office of Drug Safety, told the advisory committee.
The reporting system is considered good for spotting rare side effects, such as sudden liver failure. But it is much less reliable for distinguishing whether a drug caused a relatively common medical problem such as a heart attack. It also tends to miss risks that emerge after a few years of a drug's use.
The Los Angeles Times is a Tribune Publishing newspaper.