Back to Agenda
Session 1: LACK OF ADHERENCE AND THE CHOICE OF THE RIGHT ESTIMAND IN CLINICAL TRIALS
Session Chair(s)
Norbert Benda, PHD
Head of Biostatistics and Special Pharmacokinetics, Federal Institute For Drugs and Medical Devices (BfArM), Germany
William Malbecq, DRSC
Head of Biostatistics Europe, MSD, Belgium
Recently, an ICH concept paper on choosing appropriate estimands and defining sensitivity analyses in confirmatory clinical trials was drafted. In settings where pronounced lack of adherence to the study protocol is expected, the difference between the ideal treatment effect, if the medication is taken as directed, and the treatment effect if the medication is taken as observed is crucial in the assessment of the drug’s benefit. The handling of missing data and the use data observed after treatment discontinuation or change is strongly related to the targeted estimand. An estimand is what is being estimated and precisely defines a treatment effect regarding population, outcome measure, and the parameter defined by the underlying probabilistic model either under the assumption of perfect treatment adherence or incorporating the actual adherence. The session will discuss the use and definition of different estimands and the consequences with respect to study design and analysis in the context of the drug approval process.
Speaker(s)
Mouna Akacha, PHD
Group Head of Statistical Methodology, Novartis Pharma AG, Switzerland
Estimands and Their Role in Clinical Trials: Defining suitable primary scientific questions of interest
Michael P O'Kelly, PHD, MA
Senior Director, Centre for Statistics in Drug Development, Innovation, QuintilesIMS, Ireland
The Estimand: An ugly necessity, a pie in the sky or a useful tool?
James Henry Roger, PHD, MA
Director, Livedata (UK) Ltd, United Kingdom
Have an account?