A Knowledge Database of Physical Activity Interventions for Hypertension using Large Language Models
- 1University of Michigan
Abstract
Physical exercise lowers blood pressure, but guidelines address broad populations and rarely consider age, comorbidities, medications or baseline severity. No structured, queryable database of exercise interventions for hypertension exists, and manual curation is not scalable. We present PAHD (Physical Activity for Hypertension Database), built with large language models (LLMs). An LLM screened 2,548 PubMed records by title and abstract, then full text, selecting 255 studies. LLMs extracted each studys groups, exercise protocols, patient characteristics and results; two LLMs cross-checked the study structure, and every value carries a verbatim quote. Code standardized and joined the extractions into 17,312 records from 254 studies, one per reported comparison. Against adjudicated benchmarks, screening reached 86.3% recall and 80.6% precision, and extraction recovered 90.1% of blood pressure values with 90.5% precision. A web portal matches patients hypertension stage and comorbidities to studies. PAHD provides a structured evidence base for personalized exercise interventions on hypertension management.
Related
- Association Between Aerobic Physical Activity and Hypertension Among U.S. Adults: A Cross-Sectional Analysis of NHANES (National Health and Nutrition Examination Survey) 2021-2023
- Effects of immersive virtual reality-based physical activity interventions on anthropometric variables and physical fitness in university students: protocol for a systematic review and meta-analysis
- Physical activity and incident hypertension in relation to anthropometric and liver-related indices during 5-year follow-up in the Azar Cohort Study
- Recommendations for the development, implementation, and evaluation of technology-assisted physical activity interventions to support healthy aging: insights from PhysAgeNet
- Systematic Review and Descriptive Synthesis of Behaviour Change Techniques in Physical Activity Interventions During and Beyond Breast Cancer
BibTeX
@misc{li2026knowledge,
title = {{A Knowledge Database of Physical Activity Interventions for Hypertension using Large Language Models}},
author = {C. Li and C. Obioha and X. Cui and X. Zhou and M. Wang},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.22.26366246},
url = {https://www.medrxiv.org/content/10.64898/2026.10.22.26366246v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - A Knowledge Database of Physical Activity Interventions for Hypertension using Large Language Models
AU - C. Li
AU - C. Obioha
AU - X. Cui
AU - X. Zhou
AU - M. Wang
PY - 2026
DA - 2026/10/22
PB - medRxiv
DO - 10.64898/2026.10.22.26366246
UR - https://www.medrxiv.org/content/10.64898/2026.10.22.26366246v1
AB - Physical exercise lowers blood pressure, but guidelines address broad populations and rarely consider age, comorbidities, medications or baseline severity. No structured, queryable database of exercise interventions for hypertension exists, and manual curation is not scalable. We present PAHD (Physical Activity for Hypertension Database), built with large language models (LLMs). An LLM screened 2,548 PubMed records by title and abstract, then full text, selecting 255 studies. LLMs extracted each studys groups, exercise protocols, patient characteristics and results; two LLMs cross-checked the study structure, and every value carries a verbatim quote. Code standardized and joined the extractions into 17,312 records from 254 studies, one per reported comparison. Against adjudicated benchmarks, screening reached 86.3% recall and 80.6% precision, and extraction recovered 90.1% of blood pressure values with 90.5% precision. A web portal matches patients hypertension stage and comorbidities to studies. PAHD provides a structured evidence base for personalized exercise interventions on hypertension management.
ER -