Causal Inference with Truncation-by-Death and Unmeasured Confounding

TL;DR

This paper addresses the challenge of estimating causal effects in clinical studies with truncation by death and unmeasured confounding, proposing a new identifiable estimand and a robust estimator under certain assumptions.

Contribution

It introduces an identifiable survivor average causal effect estimator using a substitutional variable and develops a double robust AIPW estimator with proven large sample properties.

Findings

Estimator applied to stem cell transplantation data.

Proposed method achieves double robustness.

Identifiability established under monotonicity and assumptions.

Abstract

Clinical studies sometimes encounter truncation by death, rendering outcomes undefined. Statistical analysis based solely on observed survivors may give biased results because the characteristics of survivors differ between treatment groups. By principal stratification, the survivor average causal effect was proposed as a causal estimand defined in always-survivors. However, this estimand is not identifiable when there is unmeasured confounding between the treatment assignment and survival or outcome process. In this paper, we consider the comparison between an aggressive treatment and a conservative treatment with monotonicity on survival. First, we show that the survivor average causal effect on the conservative treatment is identifiable based on a substitutional variable under appropriate assumptions, even when the treatment assignment is not ignorable. Next, we propose an augmented inverse probability weighting (AIPW) type estimator for this estimand with double robustness. Finally, large sample properties of this estimator are established. The proposed method is applied to investigate the effect of allogeneic stem cell transplantation types on leukemia relapse.