Causal Inference With Outcomes Truncated by Death and Missing Not at Random
Authors:
- 19 May 2025
- journal article
- Published by Wiley in Statistics in Medicine
- Vol. 44 (10-12) , e70126
Abstract
In clinical trials, principal stratification analysis is commonly employed to address the issue of truncation by death, where a subject dies before the outcome can be measured. However, in practice, many survivor outcomes may remain uncollected or be missing not at random, posing a challenge to standard principal stratification analysis. In this article, we explore the identification, estimation, and bounds of the average treatment effect within a subpopulation of individuals who would potentially survive under both treatment and control conditions. We show that the causal parameter of interest can be identified by introducing a proxy variable that affects the outcome only through the principal strata, while requiring that the treatment variable does not directly affect the missingness mechanism. Subsequently, we propose an approach for estimating causal parameters and derive nonparametric bounds in cases where identification assumptions are violated. We illustrate the performance of the proposed method through simulation studies and a real dataset obtained from a human immunodeficiency virus study.Keywords
Funding Information
- Natural Science Foundation of Beijing Municipality (1232008)
- National Natural Science Foundation of China (12471269, 12101607, 12401378)
- National Key Research and Development Program of China (2022YFA1008100)
This publication has 44 references indexed in Scilit:
- Identification and estimation of causal effects in the presence of confounded principal strataStatistics in Medicine, 2024
- Data integration in causal inferenceWIREs Computational Statistics, 2022
- Long-acting cabotegravir plus rilpivirine for treatment in adults with HIV-1 infection: 96-week results of the randomised, open-label, phase 3 FLAIR studyThe Lancet HIV, 2021
- Semiparametric estimation of treatment effect with time-lagged response in the presence of informative censoringLifetime Data Analysis, 2011
- Mendelian randomization analysis of case‐control data using structural mean modelsStatistics in Medicine, 2010
- Covariate adjustment for two‐sample treatment comparisons in randomized clinical trials: A principled yet flexible approachStatistics in Medicine, 2008
- Attrition in longitudinal studiesJournal of Clinical Epidemiology, 2002
- Zidovudine, didanosine, and zalcitabine in the treatment of HIV infection: meta-analyses of the randomised e videnceThe Lancet, 1999
- Chapter 36 Large sample estimation and hypothesis testing1994
- A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effectMathematical Modelling, 1986