This event is hosted by the Department of Biostatistics.
PRESENTER: Mimi Kim, Sc.D. Albert Einstein College of Medicine
BIO: At the Albert Einstein College of Medicine, Dr. Kim has been the head of the Division of Biostatistics since 2003, is Director of the Biostatistics Shared Resource of the Institute of Clinical and Translational Research, and is Director of the Center for Quantitative Sciences. Her research focuses on statistical methods for designing and analyzing clinical trials and epidemiologic studies.
ABSTRACT: An increasing number of clinical trials in cancer, HIV/AIDS, and other disease areas are designed as non-inferiority trials, in which the goal is to show that the effect of an experimental therapy is not worse by more than a pre-specified margin than the standard therapy. A particularly difficult issue in these studies is that measurement error, missing data, non-compliance and other factors can obscure treatment differences and lead to anti-conservative results. In this talk, Dr. Kim will provide an overview of ongoing methodological challenges in non-inferiority trials, propose several approaches for addressing the impact of protocol deviations, and demonstrate the importance of conducting sensitivity analyses to assess robustness of results.