Semi-supervised Method for Risk Prediction with Doubly Censored EHR Data

2026-05-08

AI Summary

This study addresses the challenge of clinical risk prediction using electronic health records (EHRs), which often suffer from double truncation (both left and right truncation) and a scarcity of gold-standard labels, while commonly used surrogate outcomes may introduce substantial bias. The work proposes the first semi-supervised learning framework tailored to doubly truncated settings, integrating a small set of gold-standard labels with a large volume of surrogate outcomes subject to double truncation. By unifying survival analysis with efficient estimation theory, the method offers both theoretical guarantees and practical utility for risk prediction. Simulation studies demonstrate its significant superiority over purely supervised approaches that rely solely on labeled data. Applied to real-world EHR data, the framework successfully identifies key risk factors for type 2 diabetes, underscoring its clinical relevance and effectiveness.

Abstract

The rapid expansion of large-scale electronic health record (EHR) data offers unique opportunities to improve the accuracy and efficiency of clinical risk estimation. Yet, because clinical events may occur outside the recording health system, clinical event outcomes are frequently subject to double censoring (both left and right). Besides, gold-standard event times can often only be ascertained through labor-intensive manual chart reviews, yielding labels for only a small subset of patients. Reliance on this limited labeled set alone is limited in efficiency, whereas widely available surrogate outcomes such as the time to first diagnostic code or first disease mention are error-prone and can yield biased estimates if used directly. Semi-supervised learning (SSL) methods provide a principled way to integrate labeled and unlabeled data, and prior work has demonstrated their advantages in settings with binary or right-censored outcomes. However, existing approaches do not accommodate double censoring for risk prediction, which poses additional methodological challenges. To address this gap, we develop a novel SSL framework for risk prediction that combines a small set of gold-standard labels with large-scale surrogate information under double censoring. We establish the theoretical validity of the proposed estimator. Through extensive simulation studies, we show that our method substantially improves estimation efficiency relative to existing supervised estimators (based on the labeled data). Finally, we demonstrate its practical value by applying it to study risk factors for type 2 diabetes (T2D) using EHR data from a health system in the US.

Problem

• semi-supervised learning
• double censoring
• risk prediction
• electronic health records
• surrogate outcomes

Innovation

• semi-supervised learning
• double censoring
• risk prediction
• electronic health records
• surrogate outcomes