Robust Estimation of Heterogeneous Treatment Effects: An Algorithm-based Approach.

Pubmed ID: 38105918

Pubmed Central ID: PMC10720697

Journal: Communications in statistics: Simulation and computation

Publication Date: Jan. 1, 2023

Grants: U24 AA026969, R01 HL095086, OT2 OD026556, U2C OD023196, OT2 OD025315, OT2 OD026551, U24 OD023121, OT2 OD026552, OT2 OD026549, OT2 OD025337, OT2 OD026555, OT2 OD026553, OT2 OD023205, OT2 OD025276, OT2 OD026557, OT2 OD026554, U24 OD023163, OT2 OD023206, U24 OD023176, OT2 OD026548, OT2 OD025277, OT2 OD026550, R01 AA025208

Authors: Zhao Y, Wang H, Tu W, Li R, Su J

Cite As: Li R, Wang H, Zhao Y, Su J, Tu W. Robust Estimation of Heterogeneous Treatment Effects: An Algorithm-based Approach. Commun Stat Simul Comput 2023;52(10):4981-4998. Epub 2021 Sep 14.

Studies:

Abstract

Heterogeneous treatment effect estimation is an essential element in the practice of tailoring treatment to suit the characteristics of individual patients. Most existing methods are not sufficiently robust against data irregularities. To enhance the robustness of the existing methods, we recently put forward a general estimating equation that unifies many existing learners. But the performance of model-based learners depends heavily on the correctness of the underlying treatment effect model. This paper addresses this vulnerability by converting the treatment effect estimation to a weighted supervised learning problem. We combine the general estimating equation with supervised learning algorithms, such as the gradient boosting machine, random forest, and artificial neural network, with appropriate modifications. This extension retains the estimators' robustness while enhancing their flexibility and scalability. Simulation shows that the algorithm-based estimation methods outperform their model-based counterparts in the presence of nonlinearity and non-additivity. We developed an <b>R</b> package, <b>RCATE</b>, for public access to the proposed methods. To illustrate the methods, we present a real data example to compare the blood pressure-lowering effects of two classes of antihypertensive agents.