the basis of surrogate endpoints, and when surrogate indicators point to potential harms of a drug. The committee finds that although randomized controlled trials (RCTs) remain the gold standard for studying efficacy, there can be ethical and scientific reasons to prefer observational designs when the postmarketing research question focuses on a drug’s risks; such designs can often provide safety evidence of sufficient quality for decision-making. When requiring postmarketing RCTs, FDA has special obligations to protect patient-participants’ rights and interests, including working with relevant institutional review boards and data monitoring committees. FDA should also establish a new body to provide advice on the ethical challenges that can be posed by requiring observational studies and surveillance activities. For both ethical and scientific reasons, required postmarketing RCTs should include an accepted active treatment as at least one comparator if one is available that would probably be used if access to the drug in question were restricted.

Finally, an increased monitoring role requires that FDA establish effective interdisciplinary teams with the expertise necessary to design safety research and interpret resulting data; the necessary expertise goes beyond that necessary for drug approval. The expanded expertise includes observational study design, analysis and interpretation, Bayesian and causal inference methods, ethics, pharmacoepidemiology, outcomes research, and the design and analysis of clinical trials for safety outcomes.



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