Developing a social vulnerability index from real-world electronic health records using natural language processing
- 1University of Jyvaskyla
Abstract
BackgroundSocial vulnerability is associated with adverse health outcomes, but existing social vulnerability indices often rely on cohort-based and self-reported data, potentially limiting their population-level applicability. Therefore, this study aimed to develop an SVI from routinely collected electronic health records (EHRs) using natural language processing (NLP) and examine its association with mortality and healthcare utilization.
MethodEHR data from 193,629 individuals aged 35-102 from the Wellbeing Services County of Central Finland between years 2010 and 2023 were used in this study. A 6-item SVI (total score 0-6) was developed from free-text clinical notes using deep-learning-based NLP. SVI scores were calculated at baseline and annually for individuals. Associations between baseline SVI and all-cause mortality and healthcare utilization were assessed over 12-year follow-up using Cox and count models.
ResultsThe mean baseline SVI among all participants was 0.35 (range 0-6). Women had higher SVI (0.39) than men (0.32, p<0.001). SVI increased with age. One unit increase in SVI was associated with a higher risk of mortality (Hazard Ratio (HR) 1.23; 95% Confidence Interval (CI) (1.21-1.24), a higher odds of using healthcare services (Odds Ratio (OR) = 1.20; 95% CI): 1.19-1.21) and a greater healthcare service use (Incidence Rate Ratio (IRR) = 1.07; 95% CI: 1.06-1.07).
DiscussionAn index capturing multiple domains of social vulnerability can be derived from EHRs using NLP. Higher SVI was associated with increased mortality and healthcare utilization. These findings support the potential of EHR data for assessing social vulnerability and related health risks.
HighlightsO_LISVI can be derived from free-text EHR data using NLP in an unselected population. C_LIO_LIHigher EHR-derived social vulnerability was associated with mortality and healthcare use. C_LIO_LIEHR data may help identify individuals needing targeted assessment or support C_LI
Related
- Comorbidities of dementia in electronic health records: validation of a natural language processing app and comparison with data linkage
- Language model-assisted label refinement for accurate sepsis detection from electronic health records
- Detecting spiritual pain in electronic medical records using large language models.
- Extracting Symptoms Relevant to Geriatric Syndromes from Japanese Pharmaceutical Care Records: A Natural Language Processing Analysis
- Scalable Causal-Interpretable Machine Learning for Cancer Prescreening Using Electronic Health Records
BibTeX
@misc{tirkkonen2026developing,
title = {{Developing a social vulnerability index from real-world electronic health records using natural language processing}},
author = {A. Tirkkonen and E. Khan and J. Kaijansinkko and M. Haapanen and K. Koivunen and A. A. Peltola and M. Niemela and E. Hoogendijk and J. Lin and J. Jylhava and M. B. Von Bonsdorff},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.07.26364944},
url = {https://www.medrxiv.org/content/10.64898/2026.10.07.26364944v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Developing a social vulnerability index from real-world electronic health records using natural language processing
AU - A. Tirkkonen
AU - E. Khan
AU - J. Kaijansinkko
AU - M. Haapanen
AU - K. Koivunen
AU - A. A. Peltola
AU - M. Niemela
AU - E. Hoogendijk
AU - J. Lin
AU - J. Jylhava
AU - M. B. Von Bonsdorff
PY - 2026
DA - 2026/10/07
PB - medRxiv
DO - 10.64898/2026.10.07.26364944
UR - https://www.medrxiv.org/content/10.64898/2026.10.07.26364944v1
AB - BackgroundSocial vulnerability is associated with adverse health outcomes, but existing social vulnerability indices often rely on cohort-based and self-reported data, potentially limiting their population-level applicability. Therefore, this study aimed to develop an SVI from routinely collected electronic health records (EHRs) using natural language processing (NLP) and examine its association with mortality and healthcare utilization. MethodEHR data from 193,629 individuals aged 35-102 from the Wellbeing Services County of Central Finland between years 2010 and 2023 were used in this study. A 6-item SVI (total score 0-6) was developed from free-text clinical notes using deep-learning-based NLP. SVI scores were calculated at baseline and annually for individuals. Associations between baseline SVI and all-cause mortality and healthcare utilization were assessed over 12-year follow-up using Cox and count models. ResultsThe mean baseline SVI among all participants was 0.35 (range 0-6). Women had higher SVI (0.39) than men (0.32, p<0.001). SVI increased with age. One unit increase in SVI was associated with a higher risk of mortality (Hazard Ratio (HR) 1.23; 95% Confidence Interval (CI) (1.21-1.24), a higher odds of using healthcare services (Odds Ratio (OR) = 1.20; 95% CI): 1.19-1.21) and a greater healthcare service use (Incidence Rate Ratio (IRR) = 1.07; 95% CI: 1.06-1.07). DiscussionAn index capturing multiple domains of social vulnerability can be derived from EHRs using NLP. Higher SVI was associated with increased mortality and healthcare utilization. These findings support the potential of EHR data for assessing social vulnerability and related health risks. HighlightsO_LISVI can be derived from free-text EHR data using NLP in an unselected population. C_LIO_LIHigher EHR-derived social vulnerability was associated with mortality and healthcare use. C_LIO_LIEHR data may help identify individuals needing targeted assessment or support C_LI
ER -