With the new health care reform law, there is increasing recognition to evaluate competing pharmaceutical products directly. Statistical challenges in designing head-to-head clinical trials are not simply confined to sample size and power calculations, but include selection of critical efficacy and safety endpoints where benefit/risk can be properly evaluated and effectiveness determined. Can one such trial be sufficient to decide the comparative effectiveness of a product? What are the risks associated with a potential negative finding for a commercial product from such a trial? What can be the regulatory role in formulating a framework for such a comparative effectivenss study?