Each week, we publish layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.
Yu, J. and Lai, D. (2021), Interim analysis of sequential estimation-adjusted urn models with sample size re-estimation. Can J Statistics. https://doi.org/10.1002/cjs.11609
Clinical trials usually involve efficient and ethical objectives such as assigning more patients to better treatment and decreasing the total sample size. Different designs have been proposed to satisfy these needs. In this paper, the authors propose to combine response adaptive randomization (RAR), sequential monitoring, and sample size re-estimation in one clinical trial. RAR can achieve different objectives by skewing the assignment probabilities based on previous treatment responses and assignments. Sequential monitoring is desirable due to the following three reasons. First, it is ethical to prevent participants from being exposed to unsafe or ineffective treatments. Second, it is necessary to ensure the trial to be executed according to the protocol. Third, sequential monitoring can lead to savings in sample size, cost, and time.
In a clinical trial, a sufficient number of patients is necessary to obtain a reliable conclusion. Unfortunately, the initially planned sample size is usually not big enough due to the unpractical estimation of effect sizes. As a result, one may have to re-estimate the sample size in the middle of the trial. This paper combines the three adaptive approaches in one clinical trial and offers a comprehensive theoretical and numerical study. It has been demonstrated that the proposed methods can control the possibility of making an error and achieve efficient and ethical objectives.