Enhancing LLMs' Clinical Reasoning with Real-World Data from a Nationwide Sepsis Registry

TL;DR

This paper improves large language models' clinical reasoning by fine-tuning them with real-world sepsis registry data, leading to better performance across diverse clinical tasks and datasets.

Contribution

It introduces a method to enhance LLMs' clinical reasoning using real-world data and reinforcement learning, creating a model called C-Reason with improved capabilities.

Findings

C-Reason outperforms baseline models on clinical reasoning tasks.

The model generalizes well across different datasets and diseases.

Expert evaluations confirm improved reasoning quality.

Abstract

Although large language models (LLMs) have demonstrated impressive reasoning capabilities across general domains, their effectiveness in real-world clinical practice remains limited. This is likely due to their insufficient exposure to real-world clinical data during training, as such data is typically not included due to privacy concerns. To address this, we propose enhancing the clinical reasoning capabilities of LLMs by leveraging real-world clinical data. We constructed reasoning-intensive questions from a nationwide sepsis registry and fine-tuned Phi-4 on these questions using reinforcement learning, resulting in C-Reason. C-Reason exhibited strong clinical reasoning capabilities on the in-domain test set, as evidenced by both quantitative metrics and expert evaluations. Furthermore, its enhanced reasoning capabilities generalized to a sepsis dataset involving different tasks and patient cohorts, an open-ended consultations on antibiotics use task, and other diseases. Future research should focus on training LLMs with large-scale, multi-disease clinical datasets to develop more powerful, general-purpose clinical reasoning models.