Social determinants of health extraction from clinical notes across institutions using large language models

Detailed social determinants of health (SDoH) is often buried within clinical text in EHRs. Most current NLP efforts for SDoH have limitations, investigating limited factors, deriving data from a single institution, using specific patient cohorts/note types, with reduced focus on generalizability. We aim to address these issues by creating cross-institutional corpora and developing and evaluating the generalizability of classification models, including large language models (LLMs), for detecting SDoH factors using data from four institutions. Clinical notes were annotated with 21 SDoH factors at two levels: level 1 (SDoH factors only) and level 2 (SDoH factors and associated values). Compared to other models, instruction tuned LLM achieved top performance with micro-averaged F1 over 0.9 on level 1 corpora and over 0.84 on level 2 corpora. While models performed well when trained and tested on individual datasets, cross-dataset generalization highlighted remaining obstacles. Access to trained models will be made available at https://github.com/BIDS-Xu-Lab/LLMs4SDoH .

Authors

Institutions

Publication Details

Journal
npj Digital Medicine
Published
2025-05-17
DOI
https://doi.org/10.1038/s41746-025-01645-8
Citations
8
Primary Topic
Chronic Disease Management Strategies
Type
article
Field-Weighted Citation Impact
10.83

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Social determinants of health extraction from clinical notes across institutions using large language models

Salih Selek, Xinghan Chen, Yujia Zhou, Sunyang Fu et al.
8 citations
npj Digital Medicine
Chronic Disease Management Strategies
10.83
article

Social determinants of health extraction from clinical notes across institutions using large language models

Salih Selek, Xinghan Chen, Yujia Zhou, Sunyang Fu, Cynthia Brandt, Qingyu Chen, Jungwei Fan, Karen Wang, Cui Tao, Yifang Dang, Xinyue Hu, Vipina K. Keloth, Hua Xu, Huan He, Hongfang Liu, Christopher Gilman
article en
8 citations

Abstract

Detailed social determinants of health (SDoH) is often buried within clinical text in EHRs. Most current NLP efforts for SDoH have limitations, investigating limited factors, deriving data from a single institution, using specific patient cohorts/note types, with reduced focus on generalizability. We aim to address these issues by creating cross-institutional corpora and developing and evaluating the generalizability of classification models, including large language models (LLMs), for detecting SDoH factors using data from four institutions. Clinical notes were annotated with 21 SDoH factors at two levels: level 1 (SDoH factors only) and level 2 (SDoH factors and associated values). Compared to other models, instruction tuned LLM achieved top performance with micro-averaged F1 over 0.9 on level 1 corpora and over 0.84 on level 2 corpora. While models performed well when trained and tested on individual datasets, cross-dataset generalization highlighted remaining obstacles. Access to trained models will be made available at https://github.com/BIDS-Xu-Lab/LLMs4SDoH .

npj Digital MedicineVol. 8(1)
Yale University (US), Mayo Clinic in Florida (US), The University of Texas Health Science Center at Houston (US)
National Institute on Aging, National Human Genome Research Institute
Openalex Percentile: Top 1%
Chronic Disease Management Strategies
10.83
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.