Robust Estimation under Outcome Dependent Right Censoring in Huntington Disease: Estimators for Low and High Censoring Rates

2025-11-19

AI Summary

This paper addresses bias in standard survival models under outcome-dependent right-censoring (e.g., attrition due to disease progression in Huntington’s disease). We propose three consistent estimators: augmented inverse probability weighting (AIPW), maximum likelihood estimation (MLE), and a novel estimator tailored for high-censoring settings. Theory and simulations demonstrate that censoring rate is the key criterion for optimal estimator selection: AIPW exhibits robust performance under low-to-moderate censoring, whereas MLE suffers substantial bias under high censoring—leading to a practical, censoring-rate–based selection guideline. All estimators incorporate robust sandwich variance estimation to account for uncertainty in nuisance parameter estimation. Applied to real Huntington’s disease data, AIPW significantly reduces bias and improves confidence interval coverage, confirming estimator sensitivity to the censoring mechanism. The framework provides a generalizable, unbiased inference methodology for neurodegenerative and other diseases characterized by severe outcome-dependent censoring.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: State Estimation

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Responsible Web: Measurement, analysis, and circumvention of Web censorship
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Across health applications, researchers model outcomes as a function of time to an event, but the event time is right-censored for participants who exit the study or otherwise do not experience the event during follow-up. When censoring depends on the outcome-as in neurodegenerative disease studies where dropout is potentially related to disease severity-standard regression estimators produce biased estimates. We develop three consistent estimators for this outcome-dependent censoring setting: two augmented inverse probability weighted (AIPW) estimators and one maximum likelihood estimator (MLE). We establish their asymptotic properties and derive their robust sandwich variance estimators that account for nuisance parameter estimation. A key contribution is demonstrating that the choice of estimator to use depends on the censoring rate-the MLE performs best under low censoring rates, while the AIPW estimators yield lower bias and a higher nominal coverage under high censoring rates. We apply our estimators to Huntington disease data to characterize health decline leading up to mild cognitive impairment onset. The AIPW estimator with robustness matrix provided clinically-backed estimates with improved precision over inverse probability weighting, while MLE exhibited bias. Our results provide practical guidance for estimator selection based on censoring rate.

Problem

• Develops robust estimators for outcome-dependent right-censored data
• Addresses biased estimates when censoring depends on disease severity
• Provides selection guidance between estimators based on censoring rates

Innovation

• Developed three consistent estimators for outcome-dependent censoring
• Used AIPW estimators and maximum likelihood estimation methods
• Selected estimators based on censoring rate for optimal performance