Recent legislation requires the FDA to establish a structured framework for the regulatory assessment of the benefits and risks of new drugs. To determine whether agency reviewers had already begun embodying such a framework in their approval decisions, we examined action packages for the 19 NDAs approved August 1, 2011-July 31, 2012. We found that only eleven of the Office Director memoranda for the reviewed NDAs included explicit risk-benefit assessments, and only one used the “benefit-risk assessment framework” under discussion at FDA. With respect to the factors that reviewers use to guide their risk-benefit determinations, we found that some reviews included relatively straightforward consideration of disease severity and available therapy. In many cases, reviewers consulted competing drugs as reference points for risk and benefit assessments. In two cases, reviewers defined clinical benefit broadly, for example by factoring in the potential for use of the drug to reduce health care costs associated with negative outcomes of the indicated disease state.