Search tips
Search criteria 


Logo of bmjThis ArticleThe BMJ
BMJ. 2007 June 9; 334(7605): 1183.
PMCID: PMC1889971

New system for detecting drug side effects can beat regulatory agency

A new approach to pharmacovigilance developed by doctors in Chicago can identify adverse drug reactions up to six years before the Food and Drug Administration or monitoring programmes run by the drug industry, researchers say (Archives of Internal Medicine 2007;167:1041-9 doi: 10.1001/archinte.167.10.1041).

The scheme, the research on adverse drug events and reports (RADAR) project, was developed by Charles Bennett and colleagues at the Northwestern University Feinberg School of Medicine, in Chicago, and launched in 1998.

Dr Bennett told the BMJ that other pharmacovigilance programmes are based on epidemiology and databases. The RADAR approach is based on considering whether there is a theoretical reason why a drug might have an adverse side effect, and looking at that, he said.

The project has started 80 investigations and issued 30 reports. It focuses on incidents such as those requiring major surgery or organ transplant, and deaths. “We work actively, we are not waiting passively for all of these reports to show up in databases, we're calling people on the phone. We have a strategy to look to see if it is interesting or not and make an early decision” in terms of a full investigation.

Dr Bennett cites the example of the antiplatelet drug clopidogrel. “We were able to get that [evidence about its adverse effects] out to the FDA and into the public's hands within six months of the drug receiving approval.” Clopidogrel was approved in 1997 and the adverse reaction that the RADAR system identified was thrombotic thrombocytopenia.

“That side effect occurred at the rate of four per one million, so it is not common, though it is fatal. That would take seven years with the FDA.” Clopidogrel has a chemical structure similar to another drug with the same side effect, which allowed the doctors to form and test the risk hypothesis.

Emphasis is on the quality rather than quantity of data, with gathering and analysis carried out by a group that does not have a stake in the outcome. “We've shown with this project that people who have a basic science background and who are interested from a scientific standpoint can add a lot to the field by working with a small number of reports rather than a large number, very complete reports, and a collaborative network.”

Dr Bennett says that one weakness of the FDA reporting system is that because much of it is voluntary the data are shallow and incomplete. “I think you get what you pay for,” he says. He describes it as “essentially a bunch of statisticians waiting for the data to generate a signal.”

Raymond Woosley is president of the Critical Path Institute, based at the University of Arizona. It is a non-profit making partnership with the FDA that advances modernisation of the drug development and monitoring process.

He says that when electronic medical records are completely implemented, and that will take years, they will help the identification of adverse events from approved drugs. But that approach is never going to be completely adequate.

“We will always need programmes like RADAR, where people are trained in their community to home in and look for specific types of problems.”

Dr Woosley says, “I'm critical of Congress for not giving the FDA adequate resources and the NIH [National Institutes of Health] for not funding truly translational research that translates to people taking medicines safely.”

Articles from The BMJ are provided here courtesy of BMJ Publishing Group