Use of the Estimand Framework to Manage the Disruptive Effects of COVID-19 on Stroke Clinical Trials

Abstract

The coronavirus disease 2019 (COVID-19) pandemic has presented unique challenges to stroke care and research internationally. In particular, clinical trials in stroke are vulnerable to the impacts of the pandemic at multiple stages, including design, recruitment, intervention, follow-up, and interpretation of outcomes. A carefully considered approach is required to ensure the appropriate conduct of stroke trials during the pandemic and to maintain patient and participant safety. This has been recently addressed by the International Council for Harmonisation which, in November 2019, released an addendum to the Statistical Principles for Clinical Trials guidelines entitled Estimands and Sensitivity Analysis in Clinical Trials. In this article, we present the International Council for Harmonisation estimand framework for the design and conduct of clinical trials, with a specific focus on its application to stroke clinical trials. This framework aims to align the clinical and scientific objectives of a trial with its design and end points. It also encourages the prospective consideration of potential postrandomization intercurrent events which may occur during a trial and either impact the ability to measure an end point or its interpretation. We describe the different categories of such events and the proposed strategies for dealing with them, specifically focusing on the COVID-19 pandemic as a source of intercurrent events. We also describe potential practical impacts posed by the COVID-19 pandemic on trials, health systems, study groups, and participants, all of which should be carefully reviewed by investigators to ensure an adequate practical and statistical strategy is in place to protect trial integrity. We provide examples of the implementation of the estimand framework within hypothetical stroke trials in intracerebral hemorrhage and stroke recovery. While the focus of this article is on COVID-19 impacts, the strategies and principles proposed are well suited for other potential events or issues, which may impact clinical trials in the field of stroke.