Improving Social Determinants of Health Documentation in French EHRs Using Large Language Models

TL;DR

This study demonstrates that large language models can effectively extract social determinants of health from French clinical notes, significantly improving data completeness compared to traditional structured EHR coding.

Contribution

The paper introduces a novel LLM-based approach for extracting SDoH categories from French clinical notes, with publicly available datasets and evaluation metrics.

Findings

High accuracy for well-documented SDoH categories (F1 > 0.80)

Model identified 95.8% of patients with at least one SDoH

Performance limited by annotation inconsistencies and language-specific issues

Abstract

Social determinants of health (SDoH) significantly influence health outcomes, shaping disease progression, treatment adherence, and health disparities. However, their documentation in structured electronic health records (EHRs) is often incomplete or missing. This study presents an approach based on large language models (LLMs) for extracting 13 SDoH categories from French clinical notes. We trained Flan-T5-Large on annotated social history sections from clinical notes at Nantes University Hospital, France. We evaluated the model at two levels: (i) identification of SDoH categories and associated values, and (ii) extraction of detailed SDoH with associated temporal and quantitative information. The model performance was assessed across four datasets, including two that we publicly release as open resources. The model achieved strong performance for identifying well-documented categories such as living condition, marital status, descendants, job, tobacco, and alcohol use (F1 score > 0.80). Performance was lower for categories with limited training data or highly variable expressions, such as employment status, housing, physical activity, income, and education. Our model identified 95.8% of patients with at least one SDoH, compared to 2.8% for ICD-10 codes from structured EHR data. Our error analysis showed that performance limitations were linked to annotation inconsistencies, reliance on English-centric tokenizer, and reduced generalizability due to the model being trained on social history sections only. These results demonstrate the effectiveness of NLP in improving the completeness of real-world SDoH data in a non-English EHR system.