arXiv subjects

Pawel Morzywolek

Publications and source records attributed to Pawel Morzywolek.

4 recordsLinked to original sources

arXiv2026

Inference on Variable Importance for Treatment Effect Heterogeneity: Shapley Values and Beyond

We provide an inferential framework to assess variable importance for heterogeneous treatment effects. This assessment is especially useful in high-risk domains such as medicine, where decision makers hesitate to rely on black-box treatment recommendation algorithms. The variable importance measures we consider are local in that they may differ across individuals, while the inference is global in that it tests whether a given variable is important for any individual. Our approach builds on recent developments in semiparametric theory for function-valued parameters, and is valid even when statistical machine learning algorithms are employed to quantify treatment effect heterogeneity. We demonstrate the applicability of our method to infectious disease prevention strategies.

stat.ME↗

arXiv2024

On Weighted Orthogonal Learners for Heterogeneous Treatment Effects

Motivated by applications in personalized medicine and individualized policymaking, there is a growing interest in techniques for quantifying treatment effect heterogeneity in terms of the conditional average treatment effect (CATE). Some of the most prominent methods for CATE estimation developed in recent years are T-Learner, DR-Learner and R-Learner. The latter two were designed to improve on the former by being Neyman-orthogonal. However, the relations between them remain unclear, and likewise the literature remains vague on whether these learners converge to a useful quantity or (functional) estimand when the underlying optimization procedure is restricted to a class of functions that does not include the CATE. In this article, we provide insight into these questions by discussing DR-Learner and R-Learner as special cases of a general class of weighted Neyman-orthogonal learners for the CATE, for which we moreover derive oracle bounds. Our results shed light on how one may construct Neyman-orthogonal learners with desirable properties, on when DR-Learner may be preferred over R-Learner (and vice versa), and on novel learners that may sometimes be preferable to either of these. Theoretical findings are confirmed using results from simulation studies on synthetic data, as well as an application in critical care medicine.

stat.ME↗

arXiv2023

Handling time-dependent exposures and confounders when estimating attributable fractions -- bridging the gap between multistate and counterfactual modeling

The population-attributable fraction (PAF) expresses the proportion of events that can be ascribed to a certain exposure in a certain population. It can be strongly time-dependent because either exposure incidence or excess risk may change over time. Competing events may moreover hinder the outcome of interest from being observed. Occurrence of either of these events may, in turn, prevent the exposure of interest. Estimation approaches therefore need to carefully account for the timing of events in such highly dynamic settings. The use of multistate models has been widely encouraged to eliminate preventable yet common types of time-dependent bias. Even so, it has been pointed out that proposed multistate modeling approaches for PAF estimation fail to fully eliminate such bias. In addition, assessing whether patients die from rather than with a certain exposure not only requires adequate modeling of the timing of events but also of their confounding factors. While proposed multistate modeling approaches for confounding adjustment may adequately accommodate baseline imbalances, unlike g-methods, these proposals are not generally equipped to handle time-dependent confounding. However, the connection between multistate modeling and g-methods (e.g. inverse probability of censoring weighting) for PAF estimation is not readily apparent. In this paper, we provide a weighting-based characterization of both approaches to illustrate this connection, to pinpoint current shortcomings of multistate modeling, and to enhance intuition into simple modifications to overcome these. R code is made available to foster the uptake of g-methods for PAF estimation.

stat.ME↗

arXiv2022

On Estimation and Cross-validation of Dynamic Treatment Regimes with Competing Risks

The optimal moment to start renal replacement therapy in a patient with acute kidney injury (AKI) remains a challenging problem in intensive care nephrology. Multiple randomised controlled trials have tried to answer this question, but these can, by definition, only analyse a limited number of treatment initiation strategies. In view of this, we use routinely collected observational data from the Ghent University Hospital intensive care units (ICUs) to investigate different pre-specified timing strategies for renal replacement therapy initiation based on time-updated levels of serum potassium, pH and fluid balance in critically ill patients with AKI with the aim to minimize 30-day ICU mortality. For this purpose, we apply statistical techniques for evaluating the impact of specific dynamic treatment regimes in the presence of ICU discharge as a competing event. We discuss two approaches, a non-parametric one - using an inverse probability weighted Aalen-Johansen estimator - and a semiparametric one - using dynamic-regime marginal structural models. Furthermore, we suggest an easy to implement cross-validation technique that can be used for the out-of-sample performance assessment of the optimal dynamic treatment regime. Our work illustrates the potential of data-driven medical decision support based on routinely collected observational data.

stat.AP↗