How seating of FDA committee members might impact their votes

Food and Drug Administration (FDA) advisory committees determine patterns of medical practice in the United States and worldwide. But might the decisions of these committees be affected by where members of the committees sit — which, in turn, affects the speaking and voting order?

A new paper by Christopher L. Magee (Professor of the Practice of Engineering Systems at MIT) and Johns Hopkins University postdoc David Broniatowski addresses this question. The paper, "Does Seating Location Impact Voting Behavior on FDA Advisory Committees?", will appear next month in The American Journal of Therapeutics.

"FDA advisory panels are supposed to bring objectivity and transparency to the selection of innovative medical devices, but these panels are still human organizations. Therefore, they are subject to social dynamics that may not have been anticipated and that might not be removable," Broniatowski says.

"Someone has to speak first — and this person will likely exert some degree of influence. Instead of trying to eliminate these sources of influence, we would do better to understand how they work, and how best to use them to improve the decision-making process. This doesn't just apply to the FDA, but to any setting where panels of experts must meet to make decisions about complex technical systems."

The paper is based upon research Broniatowski conducted as part of his doctoral dissertation in the MIT Engineering Systems Division. The research was funded by the MIT Portugal Program.

Topics: Engineering Systems, Medicine, Research


FDA regulates the pharmaceutical industry, not the medical practice, which is the product of Clinical criteria of each doctor acquired from books, journals and specially the day to day experience.
CDRH changed to an electronic voting system years ago, to avoid the herding effect. With any group of people at a meeting you are always going to get more vocal participants, however, to draw a concluding between where a panelist sits and how they vote is a real stretch. The panelists sit through 2-4hrs of presentations and OPH speakers and have multiple chances to question all presenters, the panelists then have 1-4+ hrs of panel deliberations and panel questions to air out any issues from the sponsor or FDA. In addition our SGEs are highly educated and are trained to give their independent opinion of the topic at hand. Another consideration is the topic at hand, some PMAs contain better data then others, others show no risk with a minimal benefit, leaving surgeons reluctant not to have this on the market when it gives them an additional option to treat patients.
Thank you, Panel100, for your thoughtful comment. It seems that CDRH made the switch to electronic voting in 2010, as this work was being written up. Nevertheless, this issue is discussed extensively in the paper. The data show that speaking order, more than voting order, is associated with voting outcome. The order of participants' speech is strongly associated with voting outcome, even with simultaneous voting. Speaking order has historically been a function of seating location. We do not dispute that the topic at hand and training, etc., have a beneficial impact; yet, social effects, such as speaking order, still seem to carry influence over the outcome.
Back to the top