Why Are Estimands Important?
Craig H. Mallinckrodt, Geert Molenberghs, Ilya A. Lipkovich, Bohdana Ratitch · 2019
Chapter 2 describes why estimands are important. It provides a simple definition of estimands as what is to be estimated. Ambiguity in interpreting the results of medical research can arise when the target of estimation is not clearly specified in a statistical analysis plan. Estimands clarify what is to be estimated and link trial objectives with trial design and analysis and to which interventions the effect can be attributed. Clinical trials often involve multiple stakeholders who need to make different decisions. Therefore, multiple estimands are likely needed in any one trial. Moreover, perspectives on what is important to estimate may vary even for an individual stakeholder depending on the clinical context, such as stage of development, and disease state as well as to obtain a rounded view of the treatment risk–benefit profile.