• 演讲人： 张绍洪（南方科技大学）
• 时间：2020年11月26日 周四下午2:00（北京时间）
• 地点：紫金港校区行政楼1417报告厅（线下）腾讯会议 会议 ID：774 670 477
• 主办单位：浙江大学数据科学研究中心

Abstract

In long-term clinical studies, recurrent event data are sometimes collected and used to contrast the efficacies of two different treatments. The event re-occurrence rates can be compared using the popular negative binomial model, which incorporates information related to patient heterogeneity into a data analysis. In practical circumstances, patients usually accrued sequentially, thus enabling the incorporation of adaptive schemes: adjustments of the study roadmap after the assessment of the available patient responses thus far in the trial. In this talk, we focus on two valuable design methodologies for recurrent event data, adaptive in nature, that tremendously benefit both drug developers as well as trial participants. The first is response-adaptive treatment randomization and the second is group sequential design. The response-adaptive treatment allocation mechanism enables more patients to receive the superior treatment by using information gathered from cumulative outcome. For sequential monitoring, it provides a paradigm within which interim analysis can be performed in order to discover if one of the treatments is significantly more effective and hence the trial can be terminated in a much earlier stage.

报告人简介

张绍洪，教授。2019年9月加入南方科技大学。入职前，曾任香港中文大学统计系教授，博士生导师。1987年在美国乔治亚大学取得社会学及统计学硕士，其后在美国天普大学取得统计学博士。张教授曾于1998至2000年间任教于新加坡国立大学，并获该大学理学院颁发的教学奖。曾获得国家统计局第九届全国统计科学优秀成果二等奖。主要研究的领域为临床试验和医药统计，特别是自适应设计、非劣效试验、多重比较以及其在临床试验上的应用。在统计学国际期刊Annals of Statistics, Journal of the Royal Statistical society Series B, Biometrics等杂志上发表数十篇论文。

联系人：张立新 教授（stazlx@zju.edu.cn）