The analysis of causal effects when the outcome of interest is possibly truncated by death has a long history in statistics and causal inference. The survivor average causal effect is commonly identified with more assumptions than those guaranteed by the design of a randomized clinical trial. Our research demonstrates that individual level causal effects in the `always survivor' principal stratum can be quantified with no stronger identification assumptions than randomization.
Practical and causally interpretable sensitivity analysis is also developed and illustrated for discrete and continuous outcomes. Our sensitivity analysis enables statisticians to precisely and effectively incorporate clinical and regulatory judgement to make sound clinical decisions. Our methodology is the first and, as of yet, only proposed procedure that enables quantifying individual level causal effects in the presence of truncation by death and/or informative censoring using only the assumptions that are guaranteed by design of the randomized clinical trial.
|