This letter originally appeared in the Lancet on Dec 23, 2006 (vol. 368, page 2210).
18 drug advisory committees provide the US Food and Drug Administration (FDA) with independent advice. However, there have been virtually no quantitative examinations of how the committees function.
We expanded on the database from our previous study on FDA advisory committees to cover meetings between Jan 1, 1997, and June 30, 2006. Data were collected from meeting agendas and transcripts posted on the website of the FDA’s Center for Drug Evaluation and Research.
35 (24%) of 147 new molecular entities (NMEs) approved between 2000 and June 30, 2006, were preceded by advisory committee meetings (annual range 15–38%). This represents a decrease from 1998 and 1999 when 40% and 52%, respectively, of approved NMEs were preceded by meetings. NMEs with “priority” reviews (generally less than 6 months) were twice as likely to be preceded by an advisory committee meeting as NMEs approved through standard review.
For the period Jan 1, 1997, to June 30, 2006, the FDA did not make an oral scientific presentation at 18% (49/275) of drug advisory committee meetings (annual range 0–33%), but there was no trend over time. In such cases, the sponsor is allowed to make an oral presentation without a countervailing FDA presentation.
Finally, we used our previous data for 2001–04 on advisory committee vote outcomes, coded as favourable or not favourable to the drug under discussion, and then used an online catalogue of FDA-approved products to determine what regulatory decision the FDA made on the drug in the following 12 months. Advisory committee recommendations were followed by inconsistent FDA actions 20 of 71 times. Meetings with advisory committee votes favourable to the product were not more likely to be over-ridden by the FDA than those with unfavourable outcomes.
The closeness of the advisory committee vote was not a predictor of the likelihood of a subsequent inconsistent FDA action. To summarise, a declining minority of products faces advisory committees, the FDA frequently elects not to present its own interpretations of the sponsor’s data, and the rate of inconsistency between advisory committee recommendations and subsequent FDA actions is higher (28%) than is generally assumed, even though the follow-up period studied (1 year) is so short that few additional data could have been generated by the sponsor in the interim. On the basis of these analyses, the FDA is making suboptimum use of its drug advisory committee system.
We declare that we have no conflict of interest.
Asa L Tapley, Peter Lurie, Sidney M Wolfe
Public Citizen’s Health Research Group, 1600 20th Street, Washington, DC 20009, USA
 Lurie P, Almeida C, Stine N, Stine AR, Wolfe SM. Financial conflict of interest disclosure and voting patterns at food and drug administration drug advisory committee meetings. JAMA 2006; 295: 1921–28.
 US Food and Drug Administration: Center for Drug Evaluation and Research. http://www.fda.gov/oc/advisory/acdrugs.html (accessed June 6, 2006).
 Rehnquist J. FDA’s review process for new drug applications: a management review. Washington, DC: Office of Inspector General, Department of Health and Human Services, 2003. https://www.citizen.org/sites/default/files/oei-01-01-00590_0.pdf (accessed July 20, 2006).
 US Food and Drug Administration. Drugs@FDA. http://www.accessdata.fda.gov/scripts/cder/drugsatfda/index.cfm (accessed June 29, 2006).
 Pollack A. Expected ban on primatene mist raises some concerns. New York Times May 11, 2006.