Randomised clinical trials are often considered the gold standard in drug development as they are expected to be free from baseline confounding. However, randomisation does not protect from bias due to events that occur after randomisation, e.g. discontinuation of treatment, treatment switching etc.

At present, these post-randomisation events are dealt with implicitly by choices made about the data collection and statistical analysis. In particular, the ITT approach may not always lead to clinically meaningful treatment effects. In order to improve transparency and ensure alignment between trials objectives and statistical approaches, it is necessary to clearly define the treatment effect (estimand) which is to be targeted in a clinical trial. The ICH has reinforced this need by tasking a working group to develop an addendum to the main statistical guidance in drug development – the ICH E9.

This addendum is expected to be released for public consultation by mid-2017 and will likely result in a substantive evolution to traditional clinical trial design, conduct and analysis. In particular, this is a multi-disciplinary effort that requires a common understanding beyond the statistics community. In this course we will provide a non-technical introduction and encompass clinical, statistical and regulatory perspectives.