The Food and Drug Administration (FDA) is responsible for the approval of new drugs, biological products and medical devices in the United States. As part of the approval process, the FDA relies on advisory committees, which provide independent advice from outside experts. In this paper, we empirically study the process of information gathering and voting in FDA advisory committees, using data from transcripts of all meetings conducted between 2007 and 2020. To do this, we structurally estimate a dynamic model in which heterogeneous committee members hear evidence presented to them, and strategically decide when to stop gathering information and vote to recommend the approval or rejection of the product. We provide estimates of the informativeness of the evidence available to committee members, their preferences for approval and impatience, and the collective rule determining when learning stops and voting occurs. We use our estimates to compute counterfactual analysis examining changes to committee composition and institutional changes to the approval process for new products by the FDA, and how they affect the accuracy in decision-making.