Automated Auditing of Hospital Discharge Summaries for Care Transitions

2026-04-07

AI Summary

This study addresses the challenge of fragmented care and preventable readmissions caused by missing or inconsistent information in discharge summaries, a problem exacerbated by the impracticality of manual chart review at scale. For the first time, it integrates a locally deployed large language model (LLM) with the structured care transition standard DISCHARGED framework to automatically assess the presence, absence, or ambiguity of critical content in discharge summaries. Leveraging natural language processing and structured question-answering validation on the MIMIC-IV database, the approach enables scalable, automated auditing of discharge documentation quality while preserving patient privacy. The findings demonstrate the feasibility of systematic, large-scale quality improvement for clinical documentation and offer an innovative technical pathway toward standardized electronic health records.

Abstract

Incomplete or inconsistent discharge documentation is a primary driver of care fragmentation and avoidable readmissions. Despite its critical role in patient safety, auditing discharge summaries relies heavily on manual review and is difficult to scale. We propose an automated framework for large-scale auditing of discharge summaries using locally deployed Large Language Models (LLMs). Our approach operationalizes core transition-of-care requirements such as follow-up instructions, medication history and changes, patient information and clinical course, etc. into a structured validation checklist of questions based on DISCHARGED framework. Using adult inpatient summaries from the MIMIC-IV database, we utilize a privacy-preserving LLM to identify the presence, absence, or ambiguity of key documentation elements. This work demonstrates the feasibility of scalable, automated clinical auditing and provides a foundation for systematic quality improvement in electronic health record documentation.

Problem

• discharge summaries
• care transitions
• clinical auditing
• documentation quality
• readmissions

Innovation

• Large Language Models
• automated auditing
• discharge summaries
• care transitions
• privacy-preserving AI