SMART Designs and Q-learning for Dynamic Treatment Regimens

Thursday, 21 May 2015
9.30am - 10.30am
Seminar Room 2, Level 5, The Alfred Centre
99 Commercial Rd
Prahran 3004

Dynamic treatment regimens (DTRs) are sequential decision rules that specify how to adapt the type, dosage and timing of treatment according to an individual patient’s time-varying characteristics. DTRs offer a framework for operationalizing the multistage decision making in personalized clinical practice, thus providing an opportunity to improve it. They are particularly useful for the management of chronic diseases. Constructing “evidence-based” DTRs from patient data requires implementation of cutting-edge study design and analysis tools. In this talk, I will discuss the key ideas governing the paradigm of DTRs, a novel class of study designs called sequential multiple assignment randomized trial (SMART), and a regression-based analysis approach called Q-learning. The methodological developments will be illustrated through various study examples.

Dr Bibhas Chakraborty

Dr Bibhas Chakraborty

Centre for Quantitative Medicine
Duke-NUS Graduate Medical School, National University of Singapore

Dr. Bibhas Chakraborty is an Assistant Professor and Head of the Centre for Quantitative Medicine at the Duke-NUS Graduate Medical School, National University of Singapore. Prior to joining Duke-NUS, he was an Assistant Professor of Biostatistics at the Mailman School of Public Health, Columbia University, USA, where he still holds an honorary appointment. He has a masters’ degree from the Indian Statistical Institute, Calcutta, and a PhD from the Department of Statistics, University of Michigan, Ann Arbor, USA. His research interests include dynamic treatment regimens, adaptive designs, personalised medicine, bootstrap methods and statistical machine learning. He is the author of the first and only textbook on dynamic treatment regimens, published by Springer in 2013.

PDF icon Presentation slides.pdf727.15 KB