FDA Advisory Committees: One Person, One Vote

by Susan F. Wood, PhD 

FDA recently announced two draft guidances regarding advisory committees, one on public disclosure of financial conflicts of interests and the other on voting procedures. 

 

The one on disclosure of financial conflicts of interest has gotten a fair amount of attention as it begins to implement the FDA Amendments Act of 2007 (FDAAA) requirements on disclosure.  This draft guidance which calls for public disclosure 15 days in advance of any meeting of financial conflicts of interest for which a waiver is sought and granted for participation in the Advisory Committee (AC) meeting.  This guidance document follows another one released in March of 2007 limiting the participation of AC members who have financial conflicts of interest.  Under that March draft guidance (still not final), those with financial conflicts greater that $50,000 will not be allowed to participate, and those with financial conflicts less than $50,000 will only be allowed to participate as non-voting members.  This is real progress if implemented as written, and in combination with the requirement under FDAAA to begin to reduce waivers of any kind (even non-voting waivers), some of the questions about the actions of advisory committees is beginning to be addressed.

I find the second guidance issued a couple of weeks ago on voting procedures even more interesting than the one calling for more disclosure.  This guidance changes the voting procedures that are carried out in advisory committees.  The votes taken in advisory committee currently are surrounded by discussion before and during the vote and votes are made publicly in sequence with each member commenting and casting a vote in the context of the previous votes. 

Now FDA proposes a change.  From the guidance:

There has been much discussion inside and outside FDA regarding sequential versus simultaneous voting. Some have expressed concern that sequential voting, in which members cast public votes in turn, has the potential to compromise the integrity of the result.

For example, scholars and social scientists have studied the risk of âmomentum" in sequential voting, exploring whether some sequential voters may be influenced, perhaps even subconsciously, by the votes that precede theirs, especially if those votes are nearly identical or signal a clear trend[2]. This potential risk may be aggravated in the advisory committee setting, where votes are often conducted in full view of a passionate public and participatory audience. In the case of sequential voting, there is also a potential risk that comments made by a committee member or a designated federal officer (DFO) during the vote could inappropriately affect the deliberations of those who have not yet voted. Another potential risk is that comments could alter the meaning (or interpretation) of the question at issue in such a way as to cast doubt on whether all the members voted on the identical question.

In the new draft guidance, FDA seeks to put in place a system where all discussion and clarification of the question takes place prior to the vote and all votes will be done simultaneously.  After the vote, all votes will be made public and members of the committee will have an oportunity to explain or discuss their vote.

This will add both transparency and independence (for individual advisory committee members) - and both of these things are healthy for FDA Advisory Committee system.  For whatever reason that FDA is taking on these changes, these changes are in the right direction.

Happy Thanksgiving!

Susan F. Wood, PhD is Research Professor at George Washington University School of Public Health and Health Services, where she is part of the Project on Scientific Knowledge and Public Policy (SKAPP).  She also served as Director of the FDA Office of Womenâs Health from 2000-2005