CHDS Seminar with Noemi Kreif and David Glynn: Learning Optimal Treatment Rules Using Machine Learning and Decision Modelling

Abstract: New machine learning (ML) tools allow researchers to use the data to understand drivers of heterogeneous response to treatment. Using SPRINT (Systolic Blood Pressure Intervention Trial) as a case study, we outline an approach to integrate ML estimates of heterogeneous causal effects into a decision model to: 1) improve the estimation of cost and health outcomes at the population level, 2) make predictions at the individual level, and 3) construct optimal treatment allocation rules.

Dr. Kreif Bio: Noemi Kreif, PhD is a Senior Research Fellow at the Centre for Health Economics, University of York. Her research is focused on the combination of innovative causal inference and machine learning tools to provide more robust and granular evidence on the effectiveness of health policies.

Dr. Glynn Bio: David Glynn, PhD is a Research Fellow at the Centre for Health Economics, University of York. His research interests include: value of information, Bayesian methods, multimorbidity modelling, meta-epidemiology and machine learning.