Large language models to identify social determinants of health in electronic health records
This research explored the use of fine-tuned Large Language Models (LLMs) to extract social determinants of health (SDoH) from electronic health records, focusing on six key areas. The study demonstrated that these models significantly outperformed traditional methods, identifying 93.8% of patients with adverse SDoH, compared to only 2.0% by ICD-10 codes. This underscores the effectiveness of LLMs in enhancing SDoH data extraction and analysis for improved patient care and resource allocation.
.svg)