Electronic Thesis and Dissertation Repository

Thesis Format

Monograph

Degree

Doctor of Philosophy

Program

Statistics and Actuarial Sciences

Supervisor

He, Wenqing

Abstract

The statistical study of dynamic treatment regimes (DTRs) focuses on estimating sequential treatment decision rules tailored to patient-level information across multiple stages of intervention. Regression-based methods in DTR have been studied in the literature with a critical assumption that all the observed variables are precisely measured. However, this assumption is often violated in many applications. One example is the STAR*D study, in which the patient's depressive score is subject to measurement error. In this thesis, we explore problems in the context of DTR with measurement error or misclassification considered in the observed data.

The first project deals with covariate measurement error in Q-learning with continuous outcomes. The true covariate is not observable, but its replicate measurements are available in each stage. We propose a modified Q-learning algorithm with regression calibration to handle the measurement error. Given the replicate measurements, the proposed method obtains and uses the estimates of the unobserved true covariate in each stage of Q-learning.

The second project explores covariate measurement error in dynamic weighted survival modeling (DWSurv), a regression-based method dealing with survival outcomes in DTR. Internal validation data are assumed to be available with true covariates only observed in a subset of the data. Two correction methods are proposed to eliminate the effect of mismeasured covariate by obtaining the estimates of the missing true covariate in each stage of DWSurv. The consistency of the proposed estimator is established.

The third project examines Q-learning with binary outcomes being subject to misclassification. We investigate the outcome misclassification effect for internal validation data and develop a correction method to adjust for the effect in Q-learning. A probability relationship is established between the true outcome and the misclassified outcome. The estimation procedure in Q-learning is modified by including the derived probability relationship in the proposed method.

Extensive simulation studies are conducted to assess the performance of the proposed methods and to compare them with the naive method. Real data are analyzed for illustration of the proposed methods. The results showcase the importance of incorporating the errors in DTR and the competency of the proposed methods in obtaining the optimal DTR.

Summary for Lay Audience

Precision medicine is a new approach that recommends individualized treatment to a patient by taking the patient’s information into account. It differs from the traditional ‘one-size-fits-all’ clinical strategy, which ignores the patient’s heterogeneity in response to the treatment. Dynamic treatment regimes (DTRs) realize this process by providing sequential treatment decisions. However, in practice, a patient’s information that is used to infer a treatment decision often contains error-corrupted covariates or misclassified outcomes, which can be viewed as incorrect records of the patient’s characteristics or mislabeled clinical outcomes of the patient. The contaminated information may misrepresent the health status of the patients and further lead to inaccurate treatment decision-making. In this thesis, three situations are investigated in the context of DTR with error-corrupted covariates or misclassified outcomes.

The first study focuses on the problem of error-corrupted covariates in a DTR method with continuous outcomes, provided that the true covariate is not observed, but only its repeated measurements are available. The regression calibration method is employed to correct the error by using a new variable for the error-corrupted covariates, which are obtained from the available repeated measurements in the data.

The second study deals with the error-corrupted covariates in a survival-based DTR, given that the true covariate is partially observed in the data. Two correction methods are developed to correct the error-corrupted covariates. The proposed methods create estimates for the unobserved true covariate using the available error-corrupted covariate and use the created values for modeling.

The third study addresses the misclassified outcome problem in a DTR method with binary outcomes, assuming that the true outcome is only observed in a subset of data. A likelihood-based approach is proposed, which incorporates the relationship between the true outcome and misclassified outcome, through which the outcome misclassification can be corrected.

For each topic, simulation studies have demonstrated significant improvements in error correction and treatment decision-making. Real data applications have also shown the importance of including the errors in the DTR context.

Share

COinS