Optimal Dynamic Treatment Regimes Research Articles

A matching-based machine learning approach to estimating optimal dynamic treatment regimes with time-to-event outcomes.

Observational data (e.g. electronic health records) has become increasingly important in evidence-based research on dynamic treatment regimes, which tailor treatments over time to patients based on their characteristics and evolving clinical history. It is of great interest for clinicians and statisticians to identify an optimal dynamic treatment regime that can produce the best expected clinical outcome for each individual and thus maximize the treatment benefit over the population. Observational data impose various challenges for using statistical tools to estimate optimal dynamic treatment regimes. Notably, the task becomes more sophisticated when the clinical outcome of primary interest is time-to-event. Here, we propose a matching-based machine learning method to identify the optimal dynamic treatment regime with time-to-event outcomes subject to right-censoring using electronic health record data. In contrast to the established inverse probability weighting-based dynamic treatment regime methods, our proposed approach provides better protection against model misspecification and extreme weights in the context of treatment sequences, effectively addressing a prevalent challenge in the longitudinal analysis of electronic health record data. In simulations, the proposed method demonstrates robust performance across a range of scenarios. In addition, we illustrate the method with an application to estimate optimal dynamic treatment regimes for patients with advanced non-small cell lung cancer using a real-world, nationwide electronic health record database from Flatiron Health.

Multiobjective tree-based reinforcement learning for estimating tolerant dynamic treatment regimes.

A dynamic treatment regime (DTR) is a sequence of treatment decision rules that dictate individualized treatments based on evolving treatment and covariate history. It provides a vehicle for optimizing a clinical decision support system and fits well into the broader paradigm of personalized medicine. However, many real-world problems involve multiple competing priorities, and decision rules differ when trade-offs are present. Correspondingly, there may be more than one feasible decision that leads to empirically sufficient optimization. In this paper, we propose a concept of "tolerant regime," which provides a set of individualized feasible decision rules under a prespecified tolerance rate. A multiobjective tree-based reinforcement learning (MOT-RL) method is developed to directly estimate the tolerant DTR (tDTR) that optimizes multiple objectives in a multistage multitreatment setting. At each stage, MOT-RL constructs an unsupervised decision tree by modeling the counterfactual mean outcome of each objective via semiparametric regression and maximizing a purity measure constructed by the scalarized augmented inverse probability weighted estimators (SAIPWE). The algorithm is implemented in a backward inductive manner through multiple decision stages, and it estimates the optimal DTR and tDTR depending on the decision-maker's preferences. Multiobjective tree-based reinforcement learning is robust, efficient, easy-to-interpret, and flexible to different settings. We apply MOT-RL to evaluate 2-stage chemotherapy regimes that reduce disease burden and prolong survival for advanced prostate cancer patients using a dataset collected at MD Anderson Cancer Center.

Incorporating participants' welfare into sequential multiple assignment randomized trials.

Dynamic treatment regimes (DTRs) are sequences of decision rules that recommend treatments based on patients' time-varying clinical conditions. The sequential, multiple assignment, randomized trial (SMART) is an experimental design that can provide high-quality evidence for constructing optimal DTRs. In a conventional SMART, participants are randomized to available treatments at multiple stages with balanced randomization probabilities. Despite its relative simplicity of implementation and desirable performance in comparing embedded DTRs, the conventional SMART faces inevitable ethical issues, including assigning many participants to the empirically inferior treatment or the treatment they dislike, which might slow down the recruitment procedure and lead to higher attrition rates, ultimately leading to poor internal and external validities of the trial results. In this context, we propose a SMART under the Experiment-as-Market framework (SMART-EXAM), a novel SMART design that holds the potential to improve participants' welfare by incorporating their preferences and predicted treatment effects into the randomization procedure. We describe the steps of conducting a SMART-EXAM and evaluate its performance compared to the conventional SMART. The results indicate that the SMART-EXAM can improve the welfare of the participants enrolled in the trial, while also achieving a desirable ability to construct an optimal DTR when the experimental parameters are suitably specified. We finally illustrate the practical potential of the SMART-EXAM design using data from a SMART for children with attention-deficit/hyperactivity disorder.

Differentially private outcome‐weighted learning for optimal dynamic treatment regime estimation

Precision medicine is a framework for developing evidence‐based medical recommendations that seeks to determine the optimal sequence of treatments, tailored to all of the relevant, observable patient‐level characteristics. Because precision medicine relies on highly sensitive, patient‐level data, ensuring the privacy of participants is of great importance. Dynamic treatment regimes (DTRs) provide one formalization of precision medicine in a longitudinal setting. Outcome‐weighted learning (OWL) is a family of techniques for estimating optimal DTRs based on observational data. OWL techniques leverage support vector machine (SVM) classifiers in order to perform estimation. SVMs perform classification based on a set of influential points in the data known as support vectors. The classification rule produced by SVMs often requires direct access to the support vectors. Thus, releasing a treatment policy estimated with OWL requires the release of patient data for a subset of patients in the sample. As a result, the classification rules from SVMs constitute a severe privacy violation for those individuals whose data comprise the support vectors. This privacy violation is a major concern, particularly in light of the potentially highly sensitive medical data that are used in DTR estimation. Differential privacy has emerged as a mathematical framework for ensuring the privacy of individual‐level data, with provable guarantees on the likelihood that individual characteristics can be determined by an adversary. We provide the first investigation of differential privacy in the context of DTRs and provide a differentially private OWL estimator, with theoretical results allowing us to quantify the cost of privacy in terms of the accuracy of the private estimators.

Modified interactive Q-learning for attenuating the impact of model misspecification with treatment effect heterogeneity.

A sequential multiple assignment randomized trial, which incorporates multiple stages of randomization, is a popular approach for collecting data to inform personalized and adaptive treatments. There is an extensive literature on statistical methods to analyze data collected in sequential multiple assignment randomized trials and estimate the optimal dynamic treatment regime. Q-learning with linear regression is widely used for this purpose due to its ease of implementation. However, model misspecification is a common problem with this approach, and little attention has been given to the impact of model misspecification when treatment effects are heterogeneous across subjects. This article describes the integrative impact of two possible types of model misspecification related to treatment effect heterogeneity: omitted early-stage treatment effects in late-stage main effect model, and violated linearity assumption between pseudo-outcomes and predictors despite non-linearity arising from the optimization operation. The proposed method, aiming to deal with both types of misspecification concomitantly, builds interactive models into modified parametric Q-learning with Murphy's regret function. Simulations show that the proposed method is robust to both sources of model misspecification. The proposed method is applied to a two-stage sequential multiple assignment randomized trial with embedded tailoring aimed at reducing binge drinking in first-year college students.

Ambiguous Dynamic Treatment Regimes: A Reinforcement Learning Approach

A main research goal in various studies is to use an observational data set and provide a new set of counterfactual guidelines that can yield causal improvements. Dynamic Treatment Regimes (DTRs) are widely studied to formalize this process and enable researchers to find guidelines that are both personalized and dynamic. However, available methods in finding optimal DTRs often rely on assumptions that are violated in real-world applications (e.g., medical decision making or public policy), especially when (a) the existence of unobserved confounders cannot be ignored, and (b) the unobserved confounders are time varying (e.g., affected by previous actions). When such assumptions are violated, one often faces ambiguity regarding the underlying causal model that is needed to be assumed to obtain an optimal DTR. This ambiguity is inevitable because the dynamics of unobserved confounders and their causal impact on the observed part of the data cannot be understood from the observed data. Motivated by a case study of finding superior treatment regimes for patients who underwent transplantation in our partner hospital (Mayo Clinic) and faced a medical condition known as new-onset diabetes after transplantation, we extend DTRs to a new class termed Ambiguous Dynamic Treatment Regimes (ADTRs), in which the causal impact of treatment regimes is evaluated based on a “cloud” of potential causal models. We then connect ADTRs to Ambiguous Partially Observable Markov Decision Processes (APOMDPs) proposed by Saghafian (2018) , and consider unobserved confounders as latent variables but with ambiguous dynamics and causal effects on observed variables. Using this connection, we develop two reinforcement learning methods termed Direct Augmented V-Learning (DAV-Learning) and Safe Augmented V-Learning (SAV-Learning), which enable using the observed data to effectively learn an optimal treatment regime. We establish theoretical results for these learning methods, including (weak) consistency and asymptotic normality. We further evaluate the performance of these learning methods both in our case study (using clinical data) and in simulation experiments (using synthetic data). We find promising results for our proposed approaches, showing that they perform well even compared with an imaginary oracle who knows both the true causal model (of the data-generating process) and the optimal regime under that model. Finally, we highlight that our approach enables a two-way personalization; obtained treatment regimes can be personalized based on both patients’ characteristics and physicians’ preferences.This paper was accepted by David Simchi-Levi, data science.Supplemental Material: The data files and online appendix are available at https://doi.org/10.1287/mnsc.2022.00883 .

Outcome trajectory estimation for optimal dynamic treatment regimes with repeated measures.

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.

Imputation-based Q-learning for optimizing dynamic treatment regimes with right-censored survival outcome.

Q-learning has been one of the most commonly used methods for optimizing dynamic treatment regimes (DTRs) in multistage decision-making. Right-censored survival outcome poses a significant challenge to Q-Learning due to its reliance on parametric models for counterfactual estimation which are subject to misspecification and sensitive to missing covariates. In this paper, we propose an imputation-based Q-learning (IQ-learning) where flexible nonparametric or semiparametric models are employed to estimate optimal treatment rules for each stage and then weighted hot-deck multiple imputation (MI) and direct-draw MI are used to predict optimal potential survival times. Missing data are handled using inverse probability weighting and MI, and the nonrandom treatment assignment among the observed is accounted for using a propensity-score approach. We investigate the performance of IQ-learning via extensive simulations and show that it is more robust to model misspecification than existing Q-Learning methods, imputes only plausible potential survival times contrary to parametric models and provides more flexibility in terms of baseline hazard shape. Using IQ-learning, we developed an optimal DTR for leukemia treatment based on a randomized trial with observational follow-up that motivated thisstudy.

Bayesian likelihood-based regression for estimation of optimal dynamic treatment regimes

Abstract Clinicians often make sequences of treatment decisions that can be framed as dynamic treatment regimes. In this paper, we propose a Bayesian likelihood-based dynamic treatment regime model that incorporates regression specifications to yield interpretable relationships between covariates and stage-wise outcomes. We define a set of probabilistically-coherent properties for dynamic treatment regime processes and present the theoretical advantages that are consequential to these properties. We justify the likelihood-based approach by showing that it guarantees these probabilistically-coherent properties, whereas existing methods lead to process spaces that typically violate these properties and lead to modelling assumptions that are infeasible. Through a numerical study, we show that our proposed method can achieve superior performance over existing state-of-the-art methods.

Estimating individualized treatment rules in longitudinal studies with covariate-driven observation times.

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, that is, 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.

Accountable survival contrast-learning for optimal dynamic treatment regimes

Dynamic treatment regime (DTR) is an emerging paradigm in recent medical studies, which searches a series of decision rules to assign optimal treatments to each patient by taking into account individual features such as genetic, environmental, and social factors. Although there is a large and growing literature on statistical methods to estimate optimal treatment regimes, most methodologies focused on complete data. In this article, we propose an accountable contrast-learning algorithm for optimal dynamic treatment regime with survival endpoints. Our estimating procedure is originated from a doubly-robust weighted classification scheme, which is a model-based contrast-learning method that directly characterizes the interaction terms between predictors and treatments without main effects. To reflect the censorship, we adopt the pseudo-value approach that replaces survival quantities with pseudo-observations for the time-to-event outcome. Unlike many existing approaches, mostly based on complicated outcome regression modeling or inverse-probability weighting schemes, the pseudo-value approach greatly simplifies the estimating procedure for optimal treatment regime by allowing investigators to conveniently apply standard machine learning techniques to censored survival data without losing much efficiency. We further explore a SCAD-penalization to find informative clinical variables and modified algorithms to handle multiple treatment options by searching upper and lower bounds of the objective function. We demonstrate the utility of our proposal via extensive simulations and application to AIDS data.

Timing of dialysis in acute kidney injury using routinely collected data and dynamic treatment regimes

Background and objectivesDefining the optimal moment to start renal replacement therapy (RRT) in acute kidney injury (AKI) remains challenging. Multiple randomized controlled trials (RCTs) addressed this question whilst using absolute criteria such as pH or serum potassium. However, there is a need for identification of the most optimal cut-offs of these criteria. We conducted a causal analysis on routinely collected data (RCD) to compare the impact of different pre-specified dynamic treatment regimes (DTRs) for RRT initiation based on time-updated levels of potassium, pH, and urinary output on 30-day ICU mortality.Design, setting, participants, and measurementsPatients in the ICU of Ghent University Hospital were included at the time they met KDIGO-AKI-stage ≥ 2. We applied inverse-probability-of-censoring-weighted Aalen–Johansen estimators to evaluate 30-day survival under 81 DTRs prescribing RRT initiation under different thresholds of potassium, pH, or persisting oliguria.ResultsOut of 13,403 eligible patients (60.8 ± 16.8 years, SOFA 7.0 ± 4.1), 5622 (63.4 ± 15.3 years, SOFA 8.2 ± 4.2) met KDIGO-AKI-stage ≥ 2. The DTR that delayed RRT until potassium ≥ 7 mmol/l, persisting oliguria for 24–36 h, and/or pH < 7.0 (non-oliguric) or < 7.2 (oliguric) despite maximal conservative treatment resulted in a reduced 30-day ICU mortality (from 12.7% [95% CI 11.9–13.6%] under current standard of care to 10.5% [95% CI 9.5–11.7%]; risk difference 2.2% [95% CI 1.3–3.8%]) with no increase in patients starting RRT (from 471 [95% CI 430–511] to 475 [95% CI 342–572]). The fivefold cross-validation benchmark for the optimal DTR resulted in 30-day ICU mortality of 10.7%.ConclusionsOur causal analysis of RCD to compare RRT initiation at different thresholds of refractory low pH, high potassium, and persisting oliguria identified a DTR that resulted in a decrease in 30-day ICU mortality without increase in number of RRTs. Our results suggest that the current criteria to start RRT as implemented in most RCTs may be suboptimal. However, as our analysis is hypothesis generating, this optimal DTR should ideally be validated in a multicentric RCT.

Estimating Optimal Infinite Horizon Dynamic Treatment Regimes via pT-Learning

Recent advances in mobile health (mHealth) technology provide an effective way to monitor individuals’ health statuses and deliver just-in-time personalized interventions. However, the practical use of mHealth technology raises unique challenges to existing methodologies on learning an optimal dynamic treatment regime. Many mHealth applications involve decision-making with large numbers of intervention options and under an infinite time horizon setting where the number of decision stages diverges to infinity. In addition, temporary medication shortages may cause optimal treatments to be unavailable, while it is unclear what alternatives can be used. To address these challenges, we propose a Proximal Temporal consistency Learning (pT-Learning) framework to estimate an optimal regime that is adaptively adjusted between deterministic and stochastic sparse policy models. The resulting minimax estimator avoids the double sampling issue in the existing algorithms. It can be further simplified and can easily incorporate off-policy data without mismatched distribution corrections. We study theoretical properties of the sparse policy and establish finite-sample bounds on the excess risk and performance error. The proposed method is provided in our proximalDTR package and is evaluated through extensive simulation studies and the OhioT1DM mHealth dataset. Supplementary materials for this article are available online.

Penalized Spline-Involved Tree-based (PenSIT) Learning for estimating an optimal dynamic treatment regime using observational data.

Dynamic treatment regimes are a set of time-adaptive decision rules that can be used to personalize treatment across multiple stages of care. Grounded in causal inference methods, dynamic treatment regimes identify variables that differentiate the treatment effect and may be used to tailor treatments across individuals based on the patient's own characteristics - thereby representing an important step toward personalized medicine. In this manuscript we introduce Penalized Spline-Involved Tree-based Learning, which seeks to improve upon existing tree-based approaches to estimating an optimal dynamic treatment regime. Instead of using weights determined from the estimated propensity scores, which may result in unstable estimates when weights are highly variable, we predict missing counterfactual outcomes using regression models that incorporate a penalized spline of the propensity score and other covariates predictive of the outcome. We further develop a novel purity measure applied within a decision tree framework to produce a flexible yet interpretable method for estimating an optimal multi-stage multi-treatment dynamic treatment regime. In simulation experiments we demonstrate good performance of Penalized Spline-Involved Tree-based Learning relative to competing methods and, in particular, we show that Penalized Spline-Involved Tree-based Learning may be advantageous when the sample size is small and/or when the level of confounding of the outcome is high. We apply Penalized Spline-Involved Tree-based Learning to the retrospectively-collected Medical Information Mart for Intensive Care dataset to identify variables that may be used to tailor early fluid resuscitation strategies in septic patients.

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 randomized controlled trials have tried to answer this question, but these contrast only 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 prespecified 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 nonparametric 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 to assess the out-of-sample performance of the optimal dynamic treatment regime. Our work illustrates the potential of data-driven medical decision support based on routinely collected observational data.

Personalized Dynamic Treatment Regimes in Continuous Time: A Bayesian Approach for Optimizing Clinical Decisions with Timing

Accurate models of clinical actions and their impacts on disease progression are critical for estimating personalized optimal dynamic treatment regimes (DTRs) in medical/health research, especially in managing chronic conditions. Traditional statistical methods for DTRs usually focus on estimating the optimal treatment or dosage at each given medical intervention, but overlook the important question of “when this intervention should happen.” We fill this gap by developing a two-step Bayesian approach to optimize clinical decisions with timing. In the first step, we build a generative model for a sequence of medical interventions—which are discrete events in continuous time—with a marked temporal point process (MTPP) where the mark is the assigned treatment or dosage. Then this clinical action model is embedded into a Bayesian joint framework where the other components model clinical observations including longitudinal medical measurements and time-to-event data conditional on treatment histories. In the second step, we propose a policy gradient method to learn the personalized optimal clinical decision that maximizes the patient survival by interacting the MTPP with the model on clinical observations while accounting for uncertainties in clinical observations learned from the posterior inference of the Bayesian joint model in the first step. A signature application of the proposed approach is to schedule follow-up visitations and assign a dosage at each visitation for patients after kidney transplantation. We evaluate our approach with comparison to alternative methods on both simulated and real-world datasets. In our experiments, the personalized decisions made by the proposed method are clinically useful: they are interpretable and successfully help improve patient survival.

Deconfounding Actor-Critic Network with Policy Adaptation for Dynamic Treatment Regimes.

Despite intense efforts in basic and clinical research, an individualized ventilation strategy for critically ill patients remains a major challenge. Recently, dynamic treatment regime (DTR) with reinforcement learning (RL) on electronic health records (EHR) has attracted interest from both the healthcare industry and machine learning research community. However, most learned DTR policies might be biased due to the existence of confounders. Although some treatment actions non-survivors received may be helpful, if confounders cause the mortality, the training of RL models guided by long-term outcomes (e.g., 90-day mortality) would punish those treatment actions causing the learned DTR policies to be suboptimal. In this study, we develop a new deconfounding actor-critic network (DAC) to learn optimal DTR policies for patients. To alleviate confounding issues, we incorporate a patient resampling module and a confounding balance module into our actor-critic framework. To avoid punishing the effective treatment actions non-survivors received, we design a short-term reward to capture patients' immediate health state changes. Combining short-term with long-term rewards could further improve the model performance. Moreover, we introduce a policy adaptation method to successfully transfer the learned model to new-source small-scale datasets. The experimental results on one semi-synthetic and two different real-world datasets show the proposed model outperforms the state-of-the-art models. The proposed model provides individualized treatment decisions for mechanical ventilation that could improve patient outcomes.

Multi-stage optimal dynamic treatment regimes for survival outcomes with dependent censoring.

We propose a reinforcement learning method for estimating an optimal dynamic treatment regime for survival outcomes with dependent censoring. The estimator allows the failure time to be conditionally independent of censoring and dependent on the treatment decision times, supports a flexible number of treatment arms and treatment stages, and can maximize either the mean survival time or the survival probability at a certain time-point. The estimator is constructed using generalized random survival forests and can have polynomial rates of convergence. Simulations and analysis of the Atherosclerosis Risk in Communities study data suggest that the new estimator brings higher expected outcomes than existing methods in various settings.

Doubly robust estimation of optimal dynamic treatmentregimes with multicategory treatments andsurvival outcomes.

Patients with chronic diseases, such as cancer or epilepsy, are often followed through multiple stages of clinical interventions. Dynamic treatment regimes (DTRs) are sequences of decision rules that assign treatments at each stage based on measured covariates for each patient. A DTR is said to be optimal if the expectation of the desirable clinical benefit reaches a maximum when applied to a population. When there are three or more options for treatments at each decision point and the clinical outcome of interest is a time-to-event variable, estimating an optimal DTR can be complicated. We propose a doubly robust method to estimate optimal DTRs with multicategory treatments and survival outcomes. A novel blip function is defined to measure the difference in expected outcomes among treatments, and a doubly robust weighted least squares algorithm is designed for parameter estimation. Simulations using various weight functions and scenarios support the advantages of the proposed method in estimating optimal DTRs over existing approaches. We further illustrate the practical value of our method by applying it to data from the Standard and New Antiepileptic Drugs study. In this analysis, the proposed method supports the use of the new drug lamotrigine over the standard option carbamazepine. When the actual treatments match the estimated optimal treatments, survival outcomes tend to be better. The newly developed method provides a practical approach for clinicians that is not limited to cases of binary treatment options.

Generalization error bounds of dynamic treatment regimes in penalized regression-based learning

A dynamic treatment regime (DTR) is a sequence of decision rules, one per stage of intervention, that maps up-to-date patient information to a recommended treatment. Discovering an appropriate DTR for a given disease is a challenging issue especially when a large set of prognostic variables are observed. To address this problem, we propose penalized regression-based learning methods with l1 penalty to estimate the optimal DTR that would maximize the expected outcome if implemented. We also provide generalization error bounds of the estimated DTR in the setting of finite number of stages with multiple treatment options. We first examine the relationship between value and Q-functions and derive a finite sample upper bound on the difference in values between the optimal and the estimated DTRs. For practical implementation, we develop an algorithm with partial regularization via orthogonality to construct the optimal DTR. The advantages of the proposed methods are demonstrated with extensive simulation studies and data analysis of depression clinical trials.