Can we detect treatment effect waning from time-to-event data?

2025-11-24

AI Summary

This paper addresses the identifiability of treatment effect waning over time in time-to-event data. We show that, absent strong assumptions, standard survival curves cannot distinguish whether waning occurs, as selection bias and unobserved heterogeneity render the problem fundamentally non-identifiable. To overcome the limitations of the causal risk ratio (CRB)—which fails to uniquely identify the waning mechanism—we integrate principal stratification, the controlled direct effect framework, and bounding analysis to explicitly model unobserved heterogeneity. Our key contribution is a rigorous proof that reliable inference on treatment effect waning is impossible from conventional time-to-event data alone; any such inference necessarily relies on unverifiable modeling assumptions. This result carries critical implications for public health decision-making, particularly in evaluating vaccine durability and other interventions where long-term efficacy assessment is essential.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Security and Privacy: Data transparency and provenance

Abstract

Understanding how the causal effect of a treatment evolves over time, including the potential for waning, is important for informed decisions on treatment discontinuation or repetition. For example, waning vaccine protection influences booster dose recommendations, while cost-effectiveness analyses require accounting for long-term efficacy of treatments. However, there is no consensus on the methodology to assess and account for treatment effect waning. Even in randomized controlled trials, the common naïve comparison of hazard functions can lead to misleading causal conclusions due to inherent selection bias. Although comparing survival curves is sometimes recommended as a safer measure of causal effect, it only represents a cumulative effect over time and does not address treatment effect waning. We also explore recent formulations of causal hazard ratios, based on the principal stratification approach or the controlled direct effect. These causal hazard ratios cannot be identified without strong modeling assumptions, but bounds can be derived accounting for unobserved heterogeneity and one could try to use them to detect treatment effect waning. However, we illustrate that an increase in causal hazard ratios towards one does not necessarily mean that the protective effect of the treatment is fading. Furthermore, the same survival functions may correspond to both scenarios with and without waning, which shows that treatment effect waning cannot be identified from standard time-to-event data without strong untestable modeling assumptions.

Problem

• Detecting treatment effect waning from time-to-event data
• Addressing selection bias in hazard function comparisons
• Identifying causal treatment evolution without strong assumptions

Innovation

• Identifies limitations of hazard function comparisons
• Explores causal hazard ratios with principal stratification
• Demonstrates non-identifiability of waning without strong assumptions