Despite the infusion of more than $1 billion through the American Recovery and Reinvestment Act of 2009 and the potential commitment of much more money, there will never be enough funding to perform all the comparative-effectiveness studies that we want. The depth of interest in such studies became apparent when the Committee on Comparative Effectiveness Research Prioritization of the Institute of Medicine received more than 1200 nominations of distinct topics to be among the first supported under the act.1 Thus, a critical question is how to choose the studies that will be supported with substantial, but not unlimited, funds. How much should go to large database analyses, how much to medical-literature reviews, and how much to large, rigorously designed, randomized clinical trials? And what questions should such studies address?