🤖 AI Summary
This paper addresses causal survival analysis under treatment noncompliance. We propose the first multiply robust estimator to identify heterogeneous survival causal effects within principal strata (e.g., compliers, always-takers). Our method integrates the potential outcomes framework, structural nested failure time models, inverse probability weighting, and multiply robust estimation, enabling effect decomposition under principal ignorability and monotonicity assumptions. Its key contribution is the first incorporation of multiply robustness into survival analysis with noncompliance—guaranteeing consistency even if up to two of the working models (e.g., outcome, treatment assignment, or censoring models) are misspecified. Applied to the ADAPTABLE trial, our approach reveals heterogeneous effects of aspirin dosage across compliance subgroups, elucidates the mechanistic basis for a null intent-to-treat effect, informs personalized intervention strategies, and conducts sensitivity analyses for identification assumptions.
📝 Abstract
Comparative effectiveness research frequently addresses a time-to-event outcome and can require unique considerations in the presence of treatment noncompliance. Motivated by the challenges in addressing noncompliance in the ADAPTABLE pragmatic trial, we develop a multiply robust estimator to estimate the principal survival causal effects under the principal ignorability and monotonicity assumption. The multiply robust estimator involves several working models including that for the treatment assignment, the compliance strata, censoring, and time-to-event of interest. The proposed estimator is consistent even if one, and sometimes two, of the working models are misspecified. We apply the multiply robust method in the ADAPTABLE trial to evaluate the effect of low- versus high-dose aspirin assignment on patients' death and hospitalization from cardiovascular diseases. We find that, comparing to low-dose assignment, assignment to the high-dose leads to differential effects among always high-dose takers, compliers, and always low-dose takers. Such treatment effect heterogeneity contributes to the null intention-to-treatment effect, and suggests that policy makers should design personalized strategies based on potential compliance patterns to maximize treatment benefits to the entire study population. We further perform a formal sensitivity analysis for investigating the robustness of our causal conclusions under violation of two identification assumptions specific to noncompliance.
