Using off-treatment sequential multiple imputation for binary outcomes to address intercurrent events handled by a treatment policy strategy

2025-07-18

AI Summary

This study addresses missing data due to patient dropout in binary endpoints under treatment policies. We propose a post-treatment sequential multiple imputation (PT-SMI) method for causal effect estimation. Innovatively, we formulate and compare several reverse dropout models; among them, a simplified model incorporating indicator variables for concomitant events demonstrates superior robustness and practicality. Theoretical analysis and simulation studies show that this model requires at least 50% follow-up data to ensure estimation stability. Our approach integrates the treatment policy framework with observed post-discontinuation data, leveraging multiple imputation to infer the target causal effect. Applied to a Phase III rheumatoid arthritis trial, the method yields valid and interpretable estimates. This work provides a practical, empirically validated solution for handling missing data in binary-outcome treatment policy evaluation.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Game Theory and Economic Paradigms: Imperfect Information

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

The estimand framework proposes different strategies to address intercurrent events. The treatment policy strategy seems to be the most favoured as it is closely aligned with the pre-addendum intention-to-treat principle. All data for all patients should ideally be collected, however, in reality patients may withdraw from a study leading to missing data. This needs to be dealt with as part of the estimation. Several areas of research have been conducted exploring models to estimate the estimand when intercurrent events are handled using a treatment policy strategy, however the research is limited for binary endpoints. We explore different retrieved dropout models, where post-intercurrent event, the observed data can be used to multiply impute the missing post-intercurrent event data. We compare our proposed models to a simple imputation model that makes no distinction between the pre- and post-intercurrent event data, and assess varying statistical properties through a simulation study. We then provide an example how retrieved dropout models were used in practice for Phase 3 clinical trials in rheumatoid arthritis. From the models explored, we conclude that a simple retrieved dropout model including an indicator for whether or not the intercurrent event occurred is the most pragmatic choice. However, at least 50% of observed post-intercurrent event data is required for these models to work well. Therefore, the suitability of implementing this model in practice will depend on the amount of observed post-intercurrent event data available and missing data.

Problem

• Handling missing data from patient withdrawals in clinical trials
• Estimating binary outcomes under treatment policy strategy
• Comparing imputation models for post-intercurrent event data

Innovation

• Using off-treatment sequential multiple imputation
• Handling intercurrent events with treatment policy
• Simple retrieved dropout model with event indicator