A joint QoL-Survival framework with debiased estimation under truncation by death

2026-02-10

AI Summary

This study addresses the bias in quality-of-life (QoL) assessment among populations at high risk of death, where conventional methods suffer from “death truncation.” The authors propose a novel joint modeling framework that integrates QoL and survival time by extending the binary health state paradigm—common in multi-state models—to a continuous QoL trajectory, whose joint distribution with survival is characterized within a simplex space. This approach avoids extrapolating QoL beyond death, requires no strong parametric assumptions, and prevents misleading conclusions from single-effect summaries. Leveraging efficient influence functions, the method yields a semiparametric, root-n consistent estimator compatible with machine learning algorithms under clearly specified regularity conditions. Simulations and analyses of two real-world datasets demonstrate that the proposed framework delivers more robust and transparent joint evaluations of QoL and survival, particularly in high-mortality settings.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Knowledge Representation and Reasoning: Qualitative Reasoning
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Evaluating quality-of-life (QoL) outcomes in populations with high mortality risk is complicated by truncation by death, since QoL is undefined for individuals who do not survive to the planned measurement time. We propose a framework that jointly models the distribution of QoL and survival without extrapolating QoL beyond death. Inspired by multistate formulations, we extend the joint characterization of binary health states and mortality to continuous QoL outcomes. Because treatment effects cannot be meaningfully summarized in a single one-dimensional estimand without strong assumptions, our approach simultaneously considers both survival and the joint distribution of QoL and survival with the latter conveniently displayed in a simplex. We develop assumption-lean, semiparametric estimators based on efficient influence functions, yielding flexible, root-n consistent estimators that accommodate machine-learning methods while making transparent the conditions these must satisfy. The proposed method is illustrated through simulation studies and two real-data applications.

Problem

• quality of life
• truncation by death
• survival analysis
• joint modeling
• causal inference

Innovation

• truncation by death
• joint QoL-survival modeling
• semiparametric estimation
• efficient influence function
• assumption-lean inference