and postmarketing activities are that authority over postmarketing safety is solely in the hands of people who did the work of reviewing and approving a drug and that postmarketing safety activities appear to be secondary or subservient to the premarket processes and the task of approving drugs for marketing (Wolfe SM, 2006).

CDER’s culture seems to have been influenced by how premarket and postmarketing functions have been divided historically. Randomized controlled trials are the gold standard for studies leading to drug approval. Epidemiologic, population-based studies are used after approval, when a drug is on the market and being used in real-life circumstances. Medical knowledge derives from both randomized clinical trials and epidemiologic studies (including observational studies that use automated health care databases), but the methods of the two approaches differ, as does the degree of confidence that can be accorded analytic results. Although the two approaches are complementary and can both be valuable depending on the nature of the medical problem addressed—for example, population-based studies provide different kinds of information that randomized controlled studies do not deliver before approval—recent depictions of the workings of CDER suggest that the disparate respect afforded to results of the different approaches adversely affects interactions when uncertainties about the data abound and the “call” regarding regulatory action is close. Most OND reviewers are physicians who are trained to analyze prospective, randomized controlled clinical trial data, whereas the ODS/OSE staff, including the safety evaluators and the epidemiologists, must typically work with uncontrolled or observational data. Most data bearing on safety issues generated and reviewed in the postmarketing period are from case reports and from epidemiologic studies. Recent controversies show that there is sometimes a marked difference of opinion between OND and ODS/OSE about the interpretation of such data. OND staff often view observational data as “soft” and unconvincing, whereas ODS/OSE staff see them as informative and carrying great weight in evaluating postmarketing safety questions.

The interdisciplinary tension is also an obstacle to full implementation of a lifecycle approach to drug regulation, in which the preapproval process actively and creatively involves anticipation of postapproval uncertainties and a plan for addressing them. That is a clear example of how structure and culture can connect. A structure that provides opportunities for crosscutting discussion and methods—an interdisciplinary “team approach”—would go a long way to encouraging a collaborative culture, in which differing viewpoints and types of expertise can make a contribution.

In the last decade, there have been four major restructuring efforts in the variously named office responsible for postmarketing safety and in CDER; most recently, steps have been taken to clarify and elevate the previously ad hoc role of ODS/OSE as part of a broader effort to “sustain a multi-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement