An estimand is a description of the research question a trial seeks to answer, which can help researchers better understand how their study should be designed and analysed. Estimands also provide a clear way to communicate treatment effects to different stakeholders. This episode of the Trial Talk podcast features Principal Research Fellow Brennan Kahan, exploring how triallists could benefit from using estimands. Brennan also discusses his recent paper which aims to demystify new guidance on the use of estimands. Resources: •The estimands framework: a primer on the ICH E9(R1): www.bmj.com/content/384/bmj-2023-076316 •'We must let the research question drive study methods' opinion piece: www.ncbi.nlm.nih.gov/pmc/articles/PMC9908044/ •Estimands in cluster-randomized trials: choosing analyses that answer the right question: www.ncbi.nlm.nih.gov/pmc/articles/PMC9908044/

2

4/25/2024

Clinical trial results usually tell us how effective a treatment was on average for the overall group of participants, but a key question for clinicians, patients and policy makers is: which individual patients benefit most from the treatment and which don’t benefit as much? In the latest episode of the Trial Talk podcast, Peter Godolphin and David Fisher discuss a new method for determining how treatment effects differ between subgroups of patients across multiple clinical trials, as well as how other meta-analysis researchers can use it. Resources: • Estimating interactions and subgroup-specific treatment effects in meta-analysis without aggregation bias: A within-trial framework onlinelibrary.wiley.com/doi/10.1002/jrsm.1590 • Cochrane webinar recording training.cochrane.org/resource/estim…-meta-analysis • GitHub page for metafloat package in Stata github.com/UCL/metafloat • WHO REACT Group: IL6 Prospective meta-analysis jamanetwork.com/journals/jama/fullarticle/2781880 • STOPCAP c

2

3/11/2024