PaReGTA: An LLM-based EHR Data Encoding Approach to Capture Temporal Information

TL;DR

PaReGTA is a novel LLM-based framework for encoding longitudinal EHR data that captures temporal information effectively, improving classification performance in data-limited settings without extensive training.

Contribution

It introduces a new encoding method combining templated text, contrastive fine-tuning, and temporal pooling, enhancing interpretability and performance over traditional methods.

Findings

Outperforms sparse baselines in migraine classification

More stable than deep sequential models in limited data scenarios

Utilizes pre-trained LLMs for effective encoding

Abstract

Temporal information in structured electronic health records (EHRs) is often lost in sparse one-hot or count-based representations, while sequence models can be costly and data-hungry. We propose PaReGTA, an LLM-based encoding framework that (i) converts longitudinal EHR events into visit-level templated text with explicit temporal cues, (ii) learns domain-adapted visit embeddings via lightweight contrastive fine-tuning of a sentence-embedding model, and (iii) aggregates visit embeddings into a fixed-dimensional patient representation using hybrid temporal pooling that captures both recency and globally informative visits. Because PaReGTA does not require training from scratch but instead utilizes a pre-trained LLM, it can perform well even in data-limited cohorts. Furthermore, PaReGTA is model-agnostic and can benefit from future EHR-specialized sentence-embedding models. For interpretability, we introduce PaReGTA-RSS (Representation Shift Score), which quantifies clinically defined factor importance by recomputing representations after targeted factor removal and projecting representation shifts through a machine learning model. On 39,088 migraine patients from the All of Us Research Program, PaReGTA outperforms sparse baselines for migraine type classification while deep sequential models were unstable in our cohort.