Outcome trajectory estimation for optimal dynamic treatment regimes with repeated measures

Yuan Zhang; David M Vock; Megan E Patrick; Lizbeth H Finestack; Thomas A Murray

doi:10.1093/jrsssc/qlad037

Outcome trajectory estimation for optimal dynamic treatment regimes with repeated measures

J R Stat Soc Ser C Appl Stat. 2023 May 22;72(4):976-991. doi: 10.1093/jrsssc/qlad037. eCollection 2023 Aug.

Authors

Yuan Zhang¹, David M Vock², Megan E Patrick³, Lizbeth H Finestack⁴, Thomas A Murray²

Affiliations

¹ Department of Biostatistics, Epidemiology and Informatics, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA.
² Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, MN, USA.
³ Institute for Social Research, University of Michigan, Ann Arbor, MI, USA.
⁴ Department of Speech-Language-Hearing Sciences, University of Minnesota, Minneapolis, MN, USA.

Abstract

In recent sequential multiple assignment randomized trials, outcomes were assessed multiple times to evaluate longer-term impacts of the dynamic treatment regimes (DTRs). Q-learning requires a scalar response to identify the optimal DTR. Inverse probability weighting may be used to estimate the optimal outcome trajectory, but it is inefficient, susceptible to model mis-specification, and unable to characterize how treatment effects manifest over time. We propose modified Q-learning with generalized estimating equations to address these limitations and apply it to the M-bridge trial, which evaluates adaptive interventions to prevent problematic drinking among college freshmen. Simulation studies demonstrate our proposed method improves efficiency and robustness.

Keywords: Q-learning; generalized estimating equation; heterogeneous treatment effect; longitudinal outcome trajectory; sequential multiple assignment randomized trial.

Abstract

Grants and funding