Every year, medicines with dangerous side effects enter the market, but the Federal Drug Administration is unaware which medicines these are.
In a fast-paced and often light-hearted lecture yesterday, Sebastian Schneeweiss, associate professor of epidemiology at Harvard, outlined a plan to instate an alert system for potentially dangerous pharmaceuticals in front of roughly 30 graduate students and professors.
His proposed system would collect data from computer records of health care providers and correlate it to on-the-market pharmaceuticals. This would allow the FDA to identify medications with adverse effects above a level of acceptable risk.
"Right now, there's a haphazard non-system" to screen for harm once drugs hit the market, said David Savitz, professor of community health and the organizer of the seminar. If Schneeweiss's system were enacted, it could prove an efficient way of reducing the number of drugs with adverse side effects available on the market, he said.
But Schneeweiss acknowledged that before any automated system could be implemented effectively, government regulators would have to determine the maximum allowable number of adverse side effects before a drug is recalled. Currently, that level is usually only "vaguely defined," he said.
For example, Avandia, a popular diabetes drug recalled by the FDA in 2010 because of evidence that it led to heart attacks, remained on the market for 10 years because of lack of consensus about acceptable risk, according to the New York Times.
Despite the event's relatively anti-FDA tenor — Schneeweiss included several jabs at the agency's lack of rigor and emphasis on business over consumers — Savitz said he is optimistic Schneeweiss's proposal will be heard favorably.
"At the moment we're increasing pressure on health care delivery to be accountable," he said. "There's more and more of a desire to get it right. It's our job as the academic community to nudge federal agencies into doing the right thing."
Questions from the audience centered around the subjective components of drug regulation. Many explored the tension between government regulation and consumer freedom, prompting discussion as to the amount of drug information the FDA should release.
Also under scrutiny was the process for choosing which variables to account for when synthesizing large quantities of patient data.
"Despite all these great data-finding tools, there are still unknowns," Dan Escudero GS told The Herald after the lecture. "There's a lot of subjectivity in seemingly objective tools. It's interesting to see how all these things play into each other."
