ClinNoteAgents: An LLM Multi-Agent System for Predicting and Interpreting Heart Failure 30-Day Readmission from Clinical Notes

TL;DR

ClinNoteAgents leverages large language models in a multi-agent system to extract and interpret clinical notes for predicting 30-day heart failure readmissions, improving scalability and interpretability over traditional methods.

Contribution

This paper introduces ClinNoteAgents, a novel LLM-based multi-agent framework that transforms free-text clinical notes into structured data and clinician-style summaries for heart failure readmission prediction.

Findings

High extraction fidelity for clinical variables (>=90% accuracy)

Effective risk factor identification from notes

Maintains predictive signal with 60-90% text reduction

Abstract

Heart failure (HF) is one of the leading causes of rehospitalization among older adults in the United States. Although clinical notes contain rich, detailed patient information and make up a large portion of electronic health records (EHRs), they remain underutilized for HF readmission risk analysis. Traditional computational models for HF readmission often rely on expert-crafted rules, medical thesauri, and ontologies to interpret clinical notes, which are typically written under time pressure and may contain misspellings, abbreviations, and domain-specific jargon. We present ClinNoteAgents, an LLM-based multi-agent framework that transforms free-text clinical notes into (1) structured representations of clinical and social risk factors for association analysis and (2) clinician-style abstractions for HF 30-day readmission prediction. We evaluate ClinNoteAgents on 3,544 notes from 2,065 patients (readmission rate=35.16%), demonstrating high extraction fidelity for clinical variables (conditional accuracy >= 90% for multiple vitals), key risk factor identification, and preservation of predictive signal despite 60 to 90% text reduction. By reducing reliance on structured fields and minimizing manual annotation and model training, ClinNoteAgents provides a scalable and interpretable approach to note-based HF readmission risk modeling in data-limited healthcare systems.