Purpose: FDA is increasingly requested to consider over-the-counter approval of drugs (e.g., statins) to help consumers manage risk factors (e.g., cholesterol) of health conditions (e.g., coronary heart disease (CHD)). Formalizing the conditions for such approvals, including the acceptability of product labels, requires understanding consumers' ability to make acceptable choices without physician supervision. Behavioral decision research (BDR) can help FDA structure its review, integrate behavioral and risk evidence and identify evidence gaps. We apply BDR methods, retrospectively, to an FDA advisory panel's decision whether to recommend approval of an OTC low-dose statin. Method: Using evidence available at the time of FDA's decision, we first normatively model decisions facing hypothetical consumers, identifying personally-optimal choices, assuming that consumers make optimal usage of best-available risk information. We then descriptively model how well consumers will fare in identifying those choices, considering the sensitivity of choices to imperfect communications and lay inferential processes. With these models we evaluate 1) the label in terms of its impact on consumers' decision making and 2) empirical “actual-use” evidence (provided to FDA by the drug sponsor) in terms of the impact of observed choices on consumers' individual outcomes. Results: In evaluating the product label, we characterize hypothetical individuals, who upon complying with the product label would make sub-optimal decisions, either improper self-selection by people who are not within the targeted CHD risk range or improper rejection by people for whom OTC statins may effectively reduce the risk of CHD. Sub-optimal decisions were more likely for people with multiple risk factors of CHD, particularly female smokers under 55. Conclusions: We demonstrate how BDR can aid FDA in determining the acceptability of OTC drugs. Our results are consistent with the panel's judgment that OTC statins would benefit the target population of consumers and that the drug is not so potentially harmful to prevent its marketing under any circumstances. However, the product label is inadequate for consumers to make appropriate decisions about its use, particularly women under 55. We cautiously support the panel's concern that the behavioral evidence does not currently support approval of OTC statins; however we suggest that improved communication design and testing could yield conditions for future approval. We conclude with a methodology to aid policy makers in evaluating products and their communications.