The role of post intercurrent event data in the estimation of hypothetical estimands in clinical trials

Estimation of hypothetical estimands in clinical trials typically does not make use of data that may be collected after the intercurrent event (ICE). Some recent papers have shown that such data can be used for estimation of hypothetical estimands, and that statistical efficiency and power can be increased compared to using estimators that only use data before the ICE. In this paper we critically examine the efficiency and bias of estimators that do and do not exploit data collected after ICEs, in a simplified setting. We find that efficiency can only be improved by assuming certain covariate effects are common between patients who do and do not experience ICEs, and that even when such an assumption holds, gains in efficiency will typically be modest. We moreover argue that the assumptions needed to gain efficiency by using post-ICE outcomes will often not hold, such that estimators using post-ICE data may lead to biased estimates and invalid inferences. As such, we recommend that in general estimation of hypothetical estimands should be based on estimators that do not make use of post-ICE data.