Lee-Jen Wei


 

Professor Lee-Jen Wei

L.J. Wei is a professor of Biostatistics at Harvard University. Before joining Harvard, he was a professor at University of Wisconsin, University of Michigan, and George Washington University. His main research interest is in the clinical trial meth-odology, especially in design, monitoring and analysis of studies. He has developed numerous novel statistical methods which are utilized in practice. He received the prestigious Wald Medal in 2009 from the American Statistical Association for his contribution to clinical trial methodology. In January, 2014, to honor his mentorship, Harvard School of Public Health established a Wei-family scholarship to support students studying biostatis-tics. His recent research area is concentrated on the personalize medicine under the risk-benefit paradigm via biomarkers and revitalizing clinical trial methodology. Dr. Wei has been closely working with pharmaceutical industry and the regulatory agencies for developing and evaluating new drugs/devices. 

Department of Biostatistics Harvard School of Public Health Boston, MA 02115, USA

Email: wei@hsph.harvard.edu

 

Presentation



Title: Moving beyond the Comfort Zone to Practice Translational Statistics

Abstract: The primary goal for conducting a clinical study is to use efficient and reliable inference procedures to obtain robust, clinically interpretable results with respect to risk-benefit perspectives for individual patients in a well-defined target population. Unfortunately most conventional inference procedures somehow are not readily translational. We will use several examples to illustrate the issues and concerns of the conventional wisdom and then propose some trivial alternatives. For instance, meta analysis has been routinely utilized to estimate an overall treatment difference with the summary data from multiple comparative clinical studies for heterogeneous patients’ populations. The conventional fixed and random effects model procedure (and most recently, the network meta analysis) has several well-known technical issues to render non-robustness for its practical usage. More importantly, these commonly used methods do not seem following a fundamental principle of conducting a study. That is, it may not be able to identify a meaningful target population for which the resulting between-group difference summary measure can be applied with a clear clinical interpretation. In this talk, we show a simple, robust procedure based on a mixture population concept and provide a clinically meaningful between-group contrast summary for a well-defined study population. We will use the data from three recent meta analyses to illustrate the issues and concerns about the conventional procedures and also to demonstrate the new proposal under the practical setting. If time is permitted, we will discuss the issues of other inference procedures, such as the standard stratified analysis or analysis of covariance.   (Joint work with Takahiro Hasegawa, Brian Claggett, Lu Tian and Tianxi Cai.)


Click here to download slides for plenary speech