Join the Central Indiana Chapter for a virtual lunchtime meeting where our guest speaker will be Dr. Giorgos Bakoyannis from the Department of Biostatistics at the IU School of Medicine.
date: Tuesday, February 04, 2025
time: Noon - 1:00 ET
title: Estimating optimal individualized treatment rules with multistate processes
abstract:
Multistate process data are common in studies of chronic diseases such as cancer. These data are ideal for precision medicine purposes as they can be leveraged to improve more refined health outcomes, compared to standard survival outcomes, as well as incorporate patient preferences regarding quantity versus quality of life. In this work, we propose a nonparametric outcome weighted learning approach for this problem in randomized clinical trial settings. The theoretical properties of the proposed methods, including Fisher consistency and asymptotic normality of the estimated expected outcome under the estimated optimal individualized treatment rule, are rigorously established. A consistent closed-form variance estimator is provided and methodology for the calculation of simultaneous confidence intervals is proposed. Simulation studies show that the proposed methodology and inference procedures work well even with small-sample sizes and high rates of right censoring. The methodology is illustrated using data from a randomized clinical trial on the treatment of metastatic squamous-cell carcinoma of the head and neck. Finally, we present an extension of the methodology for observational data where confounding and dependent right censoring are common.
speaker:
Giorgos Bakoyannis is an Associate Professor in the Department of Biostatistics and Health Data Science at Indiana University Indianapolis, where he also serves as the Director of Public Health Science Research. His methodological research focuses on precision medicine, specifically the development of methods for estimating optimal individualized treatment rules, and causal inference. Dr. Bakoyannis’ expertise also includes the nonparametric and semiparametric analysis of complex event history data, with a particular emphasis on challenges commonly encountered in biomedical and clinical research, such as missing data, misclassification, and interval censoring. His work has been published in leading statistical and biostatistical journals, including Biometrics, Biostatistics, and Statistica Sinica. Dr. Bakoyannis has received several awards from the American Statistical Association, the International Biometric Society – Eastern North American Region, and the International Chinese Statistical Association. He has also received funding from the National Institutes of Health (NIH) as a Principal Investigator to conduct methodological research.