Share |





FDA Drug Officers: Working in the Dark
Information technology can revolutionize drug safety monitoring.

Dr. Scott Gottlieb
February 3, 2005

If Congress and critics are looking for the real Achilles heel behind the agency's recent drug safety woes, the answers don't rest with the apple-polishing agency whistle blower David Graham or the pecking order that the drug safety office occupies inside the FDA's internal bureaucratic hierarchy.

Rather, the FDA’s glaring inadequacy is the woefully outdated computer system hobbling its medical review staff.

Investors who were holding the stock of a biotech company called Neurocrine Biosciences recently got a fast education in the problems that make the drug safety issue a hard nut to crack.

The company, which develops drugs to treat neurological problems, announced it would have to re-file its application for one of its most anticipated new drugs, a sleep aid called Indiplon, with the Food and Drug Administration.

Was the application missing data? Was the company’s pivotal trial falling short? Did the drug raise any safety concerns, or fail to show that it lulled people to a restful sleep?

Nope. None of these things are true.

If you believe Neurocrine’s story, the FDA couldn’t “navigate” the company’s electronic application.

For anybody who has ever surfed the blue links inside a Google page, navigating electronic files is pretty cut and dry. But not – apparently – for the FDA. The Agency’s information technology systems are antiquated, and its ability to digest the reams of scientific data it receives with drug applications is severely hampered as a result.

The Neurocrine story will be revealed in time, but the idea of the FDA having trouble grappling with routine technologies is not beyond belief.

Although the private sector has made considerable technological progress in using IT technologies to improve drug development, the FDA has made far less progress using the same technologies to improve patient safety and mine relevant information from clinicians at the point of care.

As a result, the FDA doesn’t have the tools it needs to make better use of the increasing amount of medical data that is finding its way onto electronic medical records and inside electronic prescriptions, data that could be used to detect early signals of potential safety problems with new drugs.

One example: Many of the drugs that have been taken from the market over the last decade were withdrawn because they have caused dangerous inflammation in the liver that sometimes progressed to complete liver failure.

Not everyone will get full blown liver failure. Many patients who have these kinds of rare drug reactions will get subtle liver inflammation that will come and go, and remain largely unreported.

It’s only the severest cases that rise to the FDA’s attention. But doctors are often alerted to the subtle cases through blood tests that measure certain liver enzymes that are leaked into the blood when the liver is damaged. However, since doctors are only looking at these tests in isolation, one patient at a time, they may be inclined to dismiss the test results as harmless or ascribe elevated liver enzymes to something other than the medicine.

If the FDA had better IT tools for spotting all of these subtle cases of liver damage, it could use datamining to monitor the routine blood tests that patients started on new drugs often get. Once it saw repeated evidence of a problem, the FDA could warn doctors of potential liver damage for new drugs well before the inevitable, and possibly fatal, case of full-blown liver failure. More and more medical information on potential side-effects is accessible to the FDA through anonymous electronic medical records—provided they had the system to capture important data and use it effectively.

Until they do, the FDA will be sitting on top of a goldmine without a shovel.

Instead, the FDA relies on others to undertake the time and cost of safety monitoring, mostly doctors and drug companies who send news of potential problems to the agency. This passive reporting system often leaves the FDA at the mercy of overworked doctors to who have to fill out lengthy drug safety reports that the FDA then uses to identify and track potential drug side effects.

(The FDA even maintains some staff whose job is to read letters submitted by doctors to medical journals. The belief inside the agency is that doctors will sometimes report adverse events to the medical journals but neglect to report it to the FDA, or only report it to FDA after they have submitted it for publication.)

In short, the FDA’s current passive reporting process is slow, antiquated, expensive, and engineered only to catch the worst possible outcomes—a recipe for catastrophic drug safety failures.

So far, proposed fixes for FDA drug safety monitoring have all focused on making this antiquated system work a little faster. For example, more of the forms that doctors and manufacturers complete are now fully electronic. But doctors still have to take proactive steps to enter the information by hand.

What is needed inside the FDA are robust systems capable of proactively gathering information about the performance and safety of new drugs to complement the FDA’s passive reporting. The FDA needs datamining systems that allow it to collect more information about a drug’s use in real world, and in some cases real time, clinical practice and the capability to analyze this data and use it more effectively.

As electronic medical records and other IT systems gain wider adoption in doctor’s offices and hospitals, the FDA will have the opportunity to revolutionize drug safety for patients.

Congress should seize this opening to increase the FDA’s funding for dedicated IT improvements, rather than tinkering with an inefficient monitoring system that predates the Internet and the laptop computer.

The FDA’s problems with drug safety monitoring didn’t start with David Graham’s epiphany but with years of budget cuts and shortsighted planning that have left the agency behind the information curve, and dependent on a system that ignores the capabilities that most modern businesses take for granted.

Scott Gottlieb is a practicing physician and a fellow at the American Enterprise Institute. He recently left the FDA, where he was Director of Medical Policy Development and a senior adviser for medical technology to the FDA Commissioner.

home   spotlight   commentary   research   events   news   about   contact   links   archives
Copyright Manhattan Institute for Policy Research
52 Vanderbilt Avenue
New York, NY 10017
(212) 599-7000