Doubly Robust Learning for Estimating Individualized Treatment with Censored Data

By Y. Q. Zhao, Donglin Zeng, Eric B. Laber, Rui Song, Ming Yuan, Michael R. Kosorok. 2014

156 citations recorded by the source corpus.

Summary Individualized treatment rules recommend treatments based on individual patient characteristics in order to maximize clinical benefit. When the clinical outcome of interest is survival time, estimation is often complicated by censoring. We develop nonparametric methods for estimating an optimal individualized treatment rule in the presence of censored data. To adjust for censoring, we propose a doubly robust estimator which requires correct specification of either the censoring model or survival model, but not both; the method is shown to be Fisher consistent when either model is correct. Furthermore, we establish the convergence rate of the expected survival under the estimated optimal individualized treatment rule to the expected survival under the optimal individualized treatment rule. We illustrate the proposed methods using simulation study and data from a Phase III clinical trial on non-small cell lung cancer.

Subfields: Management Science and Operations Research, Molecular Biology, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

In Molecular Biology

In Statistics and Probability

Citations in the corpus

Learning When-to-Treat Policies (2019, methodology): “Zhao et al. (2015) provided regret bounds for nonparametric IPW-type estimators in a dynamic setting.”
Using Decision Lists to Construct Interpretable and Parsimonious Treatment Regimes (2015, background): “Examples include marginal structural mean models (Robins et al., 2008; Orellana et al., 2010); robust marginal mean models (Zhang et al., 2012); and outcomeweighted learning (Zhang et al., 2012; Zhao et al., 2012, 2015).”
Interpretable Dynamic Treatment Regimes (2016, methodology): “Our implementation of non-parametric Q-learning used the R package randomForest with default parameters settings (Liaw and Wiener, 2002). We implemented BOWL and SOWL according to the descriptions in Zhao et al. (2015). Linear kernels were used, and the amount of regularization was chosen by ve-fold”
A Bayesian Nonparametric Approach for Estimating Individualized Treatment-Response Curves (2016, methodology): “…of treatment weighted (IPTW) estimation of marginal structure models (van der Laan and Petersen, 2007), doubly robust learning (Tsiatis, 2007; Zhao et al., 2015) with applications to longitudinal data analysis (Hernán et al., 2000), survival analysis (Lunceford et al., 2002), and adaptive…”
Offline Multi-Action Policy Learning: Generalization and Optimization (2018, methodology): “n and other constants) over existing bounds of Swaminathan and Joachims (2015), Zhou et al. (2017) and Kallus (2017a). Most importantly, unlike the results of Swaminathan and Joachims (2015) and Zhou et al. (2017), our Op( 1 √ n ) bounds for CAIPWL do not require known propensities. More specificall”
Incorporating Patient Preferences into Estimation of Optimal Individualized Treatment Rules (2017, background): “…2012; Moodie et al., 2012, 2014; Laber et al., 2014; Tian et al., 2014) or policy-search-based (Orellana et al., 2010; Zhang et al., 2012, 2013; Zhao et al., 2012, 2014, 2015; Laber and Zhao, 2015), though this division 18 © 2017, The International Biometric Society is somewhat superficial…”
Learning Cost-Effective and Interpretable Treatment Regimes (2017, unknown): “Policy-search-based methods [28, 42, 41, 38, 39] search for a policy (a function which assigns treatments to individuals) within a pre-specified class of policies such that the resulting expected outcome is maximized across the population of interest.”
Precision Medicine. (2019, methodology): “For these reasons, direct search estimators based on Vn(d) work best for settings where T is small, e.g., T = 2; however, in such settings OWL and related methods can offer significant gains in terms of robustness and marginal mean outcome (see Zhang et al 2013; Zhao et al 2015a).”
A Bayesian Nonparametic Approach for Estimating Individualized Treatment-Response Curves (2016, methodology): “…of treatment weighted (IPTW) estimation of marginal structure models (van der Laan and Petersen, 2007), doubly robust learning (Tsiatis, 2007; Zhao et al., 2015) with applications to longitudinal data analysis (Hernán et al., 2000), survival analysis (Lunceford et al., 2002), and adaptive…”