Use Data-Driven Approach to Drug Analysis: Researchers

Share

Researchers from the University of Chicago Booth School of Business say if FDA were to use a more data-driven approach in its postmarket drug analysis, there would be fewer cases in which safe drugs are wrongly penalized. To make their point, the researchers used different data to reanalyze the risk of heart problems with GlaxoSmithKline’s diabetes drug Avandia (rosiglitazone).

The study reports that FDA relied in 2007 on a meta-analysis to determine that Avandia could increase the risk of heart problems. After FDA described its findings, the use of the drug dropped by 78% over 15 months. “Had FDA used a more-data driven approach to the postmarket drug surveillance,” the researchers say, “they would have found a different verdict on this drug.”

For their review of Avandia, the researchers compiled Veterans Administration data for more than 320,000 diabetes patients over eight years. They found that Avandia was not associated with major coronary events, and when patients taking Avandia were matched exactly with patients not on the drug, Avandia was associated with fewer deaths and lower mortality, and not associated with any kind of adverse coronary events.

Lead researcher John Birge says FDA is “somewhat passive” in how it collects adverse drug event data. He says that much of the adverse event information the agency receives is submitted voluntarily and cites studies indicating that 94% of adverse events are not reported.

“So, if we see maybe a few rare events that are reported,” Birge says, “FDA may think that those are more common than they actually are and may react to them more severely than they would otherwise.”

Read more