Causal effects on non-terminal event time with application to antibiotic usage and future resistance

2025-06-11

AI Summary

Evaluating the causal effect of antibiotic treatment on time-to-resistance infection faces semicompeting risks: survival status depends on treatment, rendering conventional principal stratification inapplicable for many clinically relevant patients. To address this, we propose a novel principal stratum—“Infected or Survivors” (IOS)—and define the feasible infection causal effect (FICE), enhancing clinical inclusivity. We develop an identification framework integrating a dual disease–death model with cross-world dependence parameters, and implement FICE estimation via bivariate frailty modeling, an EM algorithm, and Monte Carlo integration. We derive its large-sample bounds theoretically and validate the method on real hospital data, quantifying causal effects of distinct antibiotic regimens on time-to-resistance infection. This yields an interpretable, implementable causal assessment tool for antimicrobial resistance mitigation.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Search and Optimization: Distributed Search

Application Category

• Security and Privacy: Large-scale security measurements
• Responsible Web: Measurement, analysis, and circumvention of Web censorship
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Comparing future antibiotic resistance levels resulting from different antibiotic treatments is challenging because some patients may survive only under one of the antibiotic treatments. We embed this problem within a semi-competing risks approach to study the causal effect on resistant infection, treated as a non-terminal event time. We argue that existing principal stratification estimands for such problems exclude patients for whom a causal effect is well-defined and is of clinical interest. Therefore, we present a new principal stratum, the infected-or-survivors (ios). The ios is the subpopulation of patients who would have survived or been infected under both antibiotic treatments. This subpopulation is more inclusive than previously defined subpopulations. We target the causal effect among these patients, which we term the feasible-infection causal effect (FICE). We develop large-sample bounds under novel assumptions, and discuss the plausibility of these assumptions in our application. As an alternative, we derive FICE identification using two illness-death models with a bivariate frailty random variable. These two models are connected by a cross-world correlation parameter. Estimation is performed by an expectation-maximization algorithm followed by a Monte Carlo procedure. We apply our methods to detailed clinical data obtained from a hospital setting.

Problem

• Causal effect on resistant infection under different antibiotics
• New principal stratum for inclusive patient subpopulation analysis
• Feasible-infection causal effect estimation with novel assumptions

Innovation

• Semi-competing risks approach for causal effects
• New principal stratum: infected-or-survivors (ios)
• Illness-death models with bivariate frailty variable