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.

翻译：在临床试验中，假设性估计量的估计通常不利用干预后事件（ICE）后可能收集的数据。一些近期论文表明，此类数据可用于估计假设性估计量，且与仅使用ICE前数据的估计量相比，统计效率和检验效能可以得到提升。本文在简化设定下，批判性地考察了利用与不利用ICE后收集数据的估计量的效率与偏倚。我们发现，只有假设某些协变量效应在经历与未经历ICE的患者间具有共性时，效率才能得到改善；且即使该假设成立，效率的提升通常也较为有限。此外，我们认为，通过使用ICE后结局来提高效率所需的假设往往不成立，因此使用ICE后数据的估计量可能导致有偏估计和无效推断。有鉴于此，我们建议，假设性估计量的估计通常应基于不使用ICE后数据的估计量。