Survivor-complier effects in the presence of selection on treatment, with application to a study of prompt ICU admission

DOI10.1080/01621459.2018.1469990zbMATH Open1462.62670arXiv1704.05706OpenAlexW2608146136MaRDI QIDQ5229895FDOQ5229895

Authors: Edward H. Kennedy, Steve Harris, Luke Keele 

Publication date: 19 August 2019

Published in: Journal of the American Statistical Association (Search for Journal in Brave)

Abstract: Pre-treatment selection or censoring (`selection on treatment') can occur when two treatment levels are compared ignoring the third option of neither treatment, in `censoring by death' settings where treatment is only defined for those who survive long enough to receive it, or in general in studies where the treatment is only defined for a subset of the population. Unfortunately, the standard instrumental variable (IV) estimand is not defined in the presence of such selection, so we consider estimating a new survivor-complier causal effect. Although this effect is generally not identified under standard IV assumptions, it is possible to construct sharp bounds. We derive these bounds and give a corresponding data-driven sensitivity analysis, along with nonparametric yet efficient estimation methods. Importantly, our approach allows for high-dimensional confounding adjustment, and valid inference even after employing machine learning. Incorporating covariates can tighten bounds dramatically, especially when they are strong predictors of the selection process. We apply the methods in a UK cohort study of critical care patients to examine the mortality effects of prompt admission to the intensive care unit, using ICU bed availability as an instrument.

Full work available at URL: https://arxiv.org/abs/1704.05706

Recommendations

• Instrumental variable estimation of complier causal treatment effect with interval‐censored data
• Identification and inference for marginal average treatment effect on the treated with an instrumental variable
• Causal proportional hazards estimation with a binary instrumental variable
• Identification of Causal Effects Using Instrumental Variables
• Instrumental variable additive hazards models

zbMATH Keywords

observational studyinstrumental variablesinfluence functionpartial identificationsemiparametric theorycensoring by death

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Censored data models (62N01) Applications of statistics to biology and medical sciences; meta analysis (62P10)

Cites Work

• Asymptotic Statistics
• Instrumental variables: an econometrician's perspective
• Principal stratification in causal inference
• Identification of Causal Effects Using Instrumental Variables
• Unified methods for censored longitudinal data and causality
• Semiparametric theory and missing data.
• Title not available (Why is that?)
• Optimal Structural Nested Models for Optimal Sequential Decisions
• Think globally, act globally: an epidemiologist's perspective on instrumental variable estimation
• Using Secondary Outcomes to Sharpen Inference in Randomized Experiments With Noncompliance
• Bounds on Treatment Effects From Studies With Imperfect Compliance
• Correcting for non-compliance in randomized trials using structural nested mean models
• Marginal and nested structural models using instrumental variables
• Likelihood-Based Analysis of Causal Effects of Job-Training Programs Using Principal Stratification
• Impossibility results for nondifferentiable functionals
• Identifiability and estimation of causal effects by principal stratification with outcomes truncated by death
• Evaluating the effects of job training programs on wages through principal stratification
• Partially Identified Treatment Effects Under Imperfect Compliance: The Case of Domestic Violence
• Sharp bounds on the causal effects in randomized experiments with ``truncation-by-death
• Using Post-Outcome Measurement Information in Censoring-by-Death Problems

Cited In (6)

• Estimation of causal effects using instrumental variables with nonignorable missing covariates: application to effect of type of delivery NICU on premature infants
• Biased Encouragements and Heterogeneous Effects in an Instrumental Variable Study of Emergency General Surgical Outcomes
• Sensitivity Analysis via the Proportion of Unmeasured Confounding
• An Automated Approach to Causal Inference in Discrete Settings
• Sharp instruments for classifying compliers and generalizing causal effects
• Treatment effect bounds: an application to Swan-Ganz catheterization

This page was built for publication: Survivor-complier effects in the presence of selection on treatment, with application to a study of prompt ICU admission

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5229895)