涩里番

Title:听Doubly-Robust Bayesian Estimation of Individualized Treatment Rules Using Network Meta-Analysis.

Abstract

听

An optimal Individualized Treatment Rule (ITR) is a function that takes a patient's characteristics, such as demographics, biomarkers, and treatment history, and outputs a treatment that is expected to give the best outcome for that patient. When estimating ITRs from individual studies, power to detect important treatment-covariate interactions is often low. Additionally, all treatments of interest may not be compared head-to-head in a single study. Network Meta-Analysis (NMA) is a method of synthesizing data from multiple studies to estimate the relative effects of a set of treatments. Two-stage ITR NMA is an emerging technique for the estimation of ITRs that can improve power and simultaneously consider all relevant treatment options while protecting sensitive data. In the first stage, study-specific ITRs are estimated, and in the second stage, the study-specific ITRs are pooled using an NMA model. In this talk, we propose a doubly-robust and fully Bayesian approach to estimating study-level ITRs. We also show how missing-at-random outcomes can be accounted for in this approach. We then propose a Bayesian NMA model for pooling the study-level ITRs that leverages the full covariance matrix of the estimates. Finally, we use the methods to estimate an optimal ITR for major depressive disorder using data from three adaptive trials.

听

Speaker bio

听

Augustine Wigle currently holds an NSERC Postdoctoral Fellowship at 涩里番 supervised by Prof. Erica Moodie in the Department of Epidemiology, Biostatistics and Occupational Health. She is interested in Bayesian methods, network meta-analysis, and precision medicine. For more information, please visit: