Learning plug-in surrogate endpoints for randomized experiments

2026-05-12

AI Summary

This study addresses the challenge of identifying valid short-term surrogate endpoints when long-term outcomes are costly or infeasible to observe in randomized experiments. Existing causal criteria for surrogacy are often non-identifiable, limiting their practical utility. To overcome this, the authors propose a plug-in composite surrogate learning framework that directly optimizes the predictive performance of treatment effect estimation, thereby circumventing traditional identifiability constraints. The approach constructs surrogates from post-treatment variables and introduces two learning strategies explicitly tailored to effect prediction. Theoretical analysis demonstrates that the method yields unbiased effect estimates under standard assumptions, and empirical evaluations on both synthetic and real-world datasets show that the learned surrogates significantly outperform existing approaches in predicting the primary treatment effect.

Abstract

Surrogate endpoints are used in place of long-term outcomes in randomized experiments when observing the real outcome for a large enough cohort is prohibitively expensive or impractical. A short-term surrogate is good if the result of an experiment using the surrogate is predictive of the result of a hypothetical study using the real outcome. Much attention has been paid to formalizing this property in causal terms, but most criteria are unidentifiable and cannot be turned into practical algorithms for learning surrogate endpoints from data. To address this, we study plug-in composite surrogates, functions of post-treatment variables that may be substituted directly for the primary outcome in a randomized experiment. We propose two methods for learning plug-in surrogates that maximize effect predictiveness, and characterize the possibility of finding endpoints that yield unbiased effect estimates in representative scenarios. Finally, in both synthetic experiments with known effects and in data from a real-world experiment, we find that our method, based on directly modeling the surrogate effect, returns plug-in endpoints more predictive of the primary effect than established methods.

Problem

• surrogate endpoints
• randomized experiments
• effect predictiveness
• causal inference
• plug-in surrogates

Innovation

• plug-in surrogate
• effect predictiveness
• randomized experiments
• causal inference
• composite surrogate