-
Estimating Individualized Treatment Rules in Longitudinal Studies with Covariate-Driven Observation Times
Authors:
Janie Coulombe,
Erica E. M. Moodie,
Susan M. Shortreed,
Christel Renoux
Abstract:
The sequential treatment decisions made by physicians to treat chronic diseases are formalized in the statistical literature as dynamic treatment regimes. To date, methods for dynamic treatment regimes have been developed under the assumption that observation times, i.e., treatment and outcome monitoring times, are determined by study investigators. That assumption is often not satisfied in electr…
▽ More
The sequential treatment decisions made by physicians to treat chronic diseases are formalized in the statistical literature as dynamic treatment regimes. To date, methods for dynamic treatment regimes have been developed under the assumption that observation times, i.e., treatment and outcome monitoring times, are determined by study investigators. That assumption is often not satisfied in electronic health records data in which the outcome, the observation times, and the treatment mechanism are associated with patients' characteristics. The treatment and observation processes can lead to spurious associations between the treatment of interest and the outcome to be optimized under the dynamic treatment regime if not adequately considered in the analysis. We address these associations by incorporating two inverse weights that are functions of a patient's covariates into dynamic weighted ordinary least squares to develop optimal single stage dynamic treatment regimes, known as individualized treatment rules. We show empirically that our methodology yields consistent, multiply robust estimators. In a cohort of new users of antidepressant drugs from the United Kingdom's Clinical Practice Research Datalink, the proposed method is used to develop an optimal treatment rule that chooses between two antidepressants to optimize a utility function related to the change in body mass index.
△ Less
Submitted 19 February, 2022;
originally announced February 2022.
-
Privacy-preserving estimation of an optimal individualized treatment rule : A case study in maximizing time to severe depression-related outcomes
Authors:
Erica EM Moodie,
Janie Coulombe,
Coraline Danieli,
Christel Renoux,
Susan M Shortreed
Abstract:
Estimating individualized treatment rules - particularly in the context of right-censored outcomes - is challenging because the treatment effect heterogeneity of interest is often small, thus difficult to detect. While this motivates the use of very large datasets such as those from multiple health systems or centres, data privacy may be of concern with participating data centres reluctant to shar…
▽ More
Estimating individualized treatment rules - particularly in the context of right-censored outcomes - is challenging because the treatment effect heterogeneity of interest is often small, thus difficult to detect. While this motivates the use of very large datasets such as those from multiple health systems or centres, data privacy may be of concern with participating data centres reluctant to share individual-level data. In this case study on the treatment of depression, we demonstrate an application of distributed regression for privacy protection used in combination with dynamic weighted survival modelling (DWSurv) to estimate an optimal individualized treatment rule whilst obscuring individual-level data. In simulations, we demonstrate the flexibility of this approach to address local treatment practices that may affect confounding, and show that DWSurv retains its double robustness even when performed through a (weighted) distributed regression approach. The work is motivated by, and illustrated with, an analysis of treatment for unipolar depression using the United Kingdom's Clinical Practice Research Datalink.
△ Less
Submitted 14 February, 2022;
originally announced February 2022.
-
Estimation of the marginal effect of antidepressants on body mass index under confounding and endogenous covariate-driven monitoring times
Authors:
Janie Coulombe,
Erica E. M. Moodie,
Robert W. Platt,
Christel Renoux
Abstract:
In studying the marginal effect of antidepressants on body mass index using electronic health records data, we face several challenges. Patients' characteristics can affect the exposure (confounding) as well as the timing of routine visits (measurement process), and those characteristics may be altered following a visit which can create dependencies between the monitoring and body mass index when…
▽ More
In studying the marginal effect of antidepressants on body mass index using electronic health records data, we face several challenges. Patients' characteristics can affect the exposure (confounding) as well as the timing of routine visits (measurement process), and those characteristics may be altered following a visit which can create dependencies between the monitoring and body mass index when viewed as a stochastic or random processes in time. This may result in a form of selection bias that distorts the estimation of the marginal effect of the antidepressant. Inverse intensity of visit weights have been proposed to adjust for these imbalances, however no approaches have addressed complex settings where the covariate and the monitoring processes affect each other in time so as to induce endogeneity, a situation likely to occur in electronic health records. We review how selection bias due to outcome-dependent follow-up times may arise and propose a new cumulated weight that models a complete monitoring path so as to address the above-mentioned challenges and produce a reliable estimate of the impact of antidepressants on body mass index. More specifically, we do so using data from the Clinical Practice Research Datalink in the United Kingdom, comparing the marginal effect of two commonly used antidepressants, citalopram and fluoxetine, on body mass index. The results are compared to those obtained with simpler methods that do not account for the extent of the dependence due to an endogenous covariate process.
△ Less
Submitted 27 June, 2021;
originally announced June 2021.
