Generalized G-estimation and Model Selection

Abstract

Dynamic treatment regimes (DTRs) aim to formalize personalized medicine by tailoring treatment decisions to individual patient characteristics. G-estimation for DTR identification targets the parameters of a structural nested mean model known as the blip function from which the optimal DTR is derived. Despite considerable work deriving such estimation methods, there has been little focus on extending G-estimation to the case of non-additive effects, non-continuous outcomes or on model selection. We demonstrate how G-estimation can be more widely applied through the use of iteratively-reweighted least squares procedures, and illustrate this for log-linear models. We then derive a quasi-likelihood function for G-estimation within the DTR framework, and show how it can be used to form an information criterion for blip model selection. These developments are demonstrated through application to a variety of simulation studies as well as data from the Sequenced Treatment Alternatives to Relieve Depression study.