On the achievability of efficiency bounds for covariate-adjusted response-adaptive randomization

Statistical Methods in Medical Research, Journal Year: 2025, Volume and Issue: unknown

Published: March 31, 2025

In the context of precision medicine, covariate-adjusted response-adaptive (CARA) randomization has garnered much attention from both academia and industry due to its benefits in providing ethical tailored treatment assignments based on patients’ profiles while still preserving favorable statistical properties. Recent years have seen substantial progress inference for various adaptive experimental designs. particular, research focused two important perspectives: how obtain robust presence model misspecification, what smallest variance, i.e., efficiency bound, an estimator can achieve. Notably, Armstrong (2022) derived asymptotic bound any procedure that assigns treatments depending covariates accrued responses, thus including CARA, among others. However, best our knowledge, no existing literature addressed whether this be achieved under CARA. paper, by connecting strands literature, namely we provide a definitive answer practical scenario where only discrete are observed used stratification. We consider special type stratified version doubly-adaptive biased coin design prove difference-in-means achieves (2022)’s with possible constraints assignments. Our work provides new insights demonstrates potential more CARA designs maximize adhering considerations. Future studies could explore achieving continuous covariates, which remains open question.

Language: Английский

---

Covariate selection for optimizing balance with an innovative adaptive randomization approach

Statistical Methods in Medical Research, Journal Year: 2025, Volume and Issue: unknown

Published: April 13, 2025

Balancing influential covariates is crucial for valid treatment comparisons in clinical studies. While covariate-adaptive randomization commonly used to achieve balance, its performance can be inadequate when the number of baseline large. It is, therefore, essential identify factors associated with outcome and ensure balance among these critical covariates. In this article, we propose a novel adaptive approach that integrates patients' responses information select sequentially significant maintain their balance. We establish theoretically consistency our covariate selection method demonstrate improved balancing, as evidenced by faster convergence rate imbalance measure, leads higher efficiency estimating effects. Furthermore, provide extensive numerical empirical studies illustrate benefits proposed across various settings.

Language: Английский

---

Covariate-Adjusted Response Adaptive Designs for Competing Risk Survival Models

Statistics in Biopharmaceutical Research, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 38

Published: Dec. 27, 2024

Often in medical research, response to a particular treatment can be classified terms of failure from multiple causes. In such cases, competing event precludes the observation main interest. Such scenarios survival analysis are termed as risks. Covariate-adjusted response-adaptive (CARA) designs skew patient allocation towards better-performing arm, so far clinical trial, for given patient's covariate profile. When there events occurring ignoring information during design stage may bias results disease-specific comparison. Optimal CARA developed, assuming proportional sub-distribution hazards two-arm trials, where primary endpoint encounters The derived proportions targeted using biased coin procedure. These that sequentially estimated, converge empirically expected target values, which functions Fine and Gray (1999) model coefficients. proposed methods shown suitable alternatives traditional balanced through extensive simulation studies have also been implemented re-design real-life trial. Simulation reveal need theoretical procedure more complicated semi-parametric models.

Language: Английский

---