[Statlist] Announcement: BBS Webinar "Controlling the chances of false discoveries in exploratory analysis of clinical trials" on 29th August 2024 (14 - 17h CET)

[Blatna, Bibiana]

Dear BBS colleagues,
We are pleased to announce the BBS Webinar "Controlling the chances of false discoveries in exploratory analysis of clinical trials" on 29th August 2024.
This virtual event is free of costs, but registration is mandatory.
Please follow the links to: AGENDA<https://baselbiometrics.github.io/home/docs/upcoming/20240829/agenda.pdf> and REGISTRATION<https://docs.google.com/forms/d/e/1FAIpQLSe5dfKHdN6JdKAFsE8Zw1CUs7soHzgrZb9-YCLGa4PRcxWoBQ/viewform>.
Kind Regards,
Organizers: Kostas Sechidis (Novartis), Frank Bretz (Novartis)

ABSTRACT
While the primary focus of clinical trials is to estimate causal effects, the collected data can also be invaluable for additional research, such as identifying variables and/or groups of
patients with desirable characteristics. Some common exploratory analysis activities focus on using clinical trial data for variable selection. For example, we may want to identify baseline
variables that are strongly associated with the disease outcome, irrespective of the treatment assignment (i.e., prognostic variables) or baseline variables that influence the treatment effect
(i.e., predictive variables). Clinical trial data can also be used for subgroup discovery, where, for example, we aim to identify groups of patients that experience a significant treatment
effect. In all these selection problems, it is critical to control the chances of false discoveries (type-I errors) to provide guarantees concerning the replicability of our results. The focus of
this session is on recent methodologies for performing this type of selection by controlling the type-I error rate.



	[[alternative HTML version deleted]]