报告名称：Response-adaptive trial design with ordinal responses

报告专家：钟俊江

专家所在单位：厦门理工学院

报告时间：2021年10月28日：15：30-16：30

报告地点：数学与统计学学院201学术报告厅

专家简介：钟俊江，厦门理工学院应用数学系副教授。2017年获台湾地区成功大学统计学博士学位，后在香港中文大学统计系从事博士后研究，曾多次受邀赴香港中文大学进行学术访问。主要从事临床试验、生物医学统计等方面的研究工作，主持教育部人文社科青年基金1项。曾在《StatisticsinMedicine》、《Statistical Methods in Medical Research》及其他国际统计期刊上发表多篇论文。

报告摘要：Ordinal responses are common in clinical studies. Although the proportional odds model is a popular option for analyzing ordered-categorical data, it cannot control the type I error rate when the proportional odds assumption fails to hold. Recently, the latent variable model was proposed to remedy the problem. The latent normal model performs very well in general, but when the data are highly skewed, the latent Weibull model is superior to the latent normal model. In this talk, we propose a response-adaptive trial design based on the latent Weibull model, which enables adjustments to be made to the study roadmap during the trial, after the assessment of patient responses to treatment. The design scheme incorporates two adaptive-design methodologies that are of enormous benefit to drug developers and trial participants: response-adaptive treatment randomization and group sequential design. We perform a simulation study that clearly demonstrates the merits of our proposed framework. Furthermore, a clinical example is used to illustrate the advantages of our response-adaptive approach.

邀请人：刘展

（审核：郑大彬）