Proceedings 2nd ISI Regional Statistics Conference, 20-24 March 2017, Indonesia (Session IPS26)
Estimating Optimal Dynamic Treatment Regimes with Shared Decision Rules
Bibhas Chakraborty*
Duke-National University of Singapore Medical School, Singapore –
[email protected]
Abstract
A dynamic treatment regime consists of decision rules that recommend how to individualize treatment
to patients based on available treatment and covariate history. In many scientific domains, these decision
rules are shared across stages of intervention. Estimating these shared decision rules often amounts to
estimating parameters indexing the decision rules that are shared across stages. In this talk, we will
present a novel simultaneous estimation procedure for the shared parameters based on Q-learning.
Through an extensive simulation study, we will present the merit of the proposed method over
reasonable competitors, in terms of the treatment allocation matching of the procedure with an “oracle”
procedure, defined as the one that makes treatment recommendations based on the true parameter values
as opposed to their estimates. We will also look at bias and mean squared error of the individual
parameter-estimates as secondary metrics. We will illustrate the proposed methodology by analyzing
data from STAR*D, a multistage randomized clinical trial for treating major depression. (This is based
on joint work with Palash Ghosh, Erica E.M. Moodie, and A. John Rush.)