Phase 1 Implementation of LLM-generated Discharge Summaries showing high Adoption in a Dutch Academic Hospital

arXiv cs.CL Papers

Summary

A 9-week pilot at a Dutch academic hospital shows 58% of admissions used LLM-generated discharge drafts, with 87% of clinicians reporting reduced documentation time and 91% intending continued use.

arXiv:2604.19774v1 Announce Type: new Abstract: Writing discharge summaries to transfer medical information is an important but time-consuming process that can be assisted by Large Language Models (LLMs). This prospective mixed methods pilot study evaluated an Electronic Health Record (EHR)-integrated LLM to generate discharge summaries drafts. In total, 379 discharge summaries were generated in clinical practice by 21 residents and 4 physician assistants during 9 weeks in our academic hospital. LLM-generated text was copied in 58.5% of admissions, and identifiable LLM content could be traced to 29.1% of final discharge letters. Notably, 86.9% of users self-reported a reduction in documentation time, and 60.9% a reduction in administrative workload. Intent to use after the pilot phase was high (91.3%), supporting further implementation of this use-case. Accurately measuring the documentation time of users on discharge summaries remains challenging, but will be necessary for future extrinsic evaluation of LLM-assisted documentation.
Original Article
View Cached Full Text

Cached at: 04/23/26, 10:02 AM

# Phase 1 Implementation of LLM-generated Discharge Summaries showing high Adoption in a Dutch Academic Hospital
Source: [https://arxiv.org/abs/2604.19774](https://arxiv.org/abs/2604.19774)
[View PDF](https://arxiv.org/pdf/2604.19774)

> Abstract:Writing discharge summaries to transfer medical information is an important but time\-consuming process that can be assisted by Large Language Models \(LLMs\)\. This prospective mixed methods pilot study evaluated an Electronic Health Record \(EHR\)\-integrated LLM to generate discharge summaries drafts\. In total, 379 discharge summaries were generated in clinical practice by 21 residents and 4 physician assistants during 9 weeks in our academic hospital\. LLM\-generated text was copied in 58\.5% of admissions, and identifiable LLM content could be traced to 29\.1% of final discharge letters\. Notably, 86\.9% of users self\-reported a reduction in documentation time, and 60\.9% a reduction in administrative workload\. Intent to use after the pilot phase was high \(91\.3%\), supporting further implementation of this use\-case\. Accurately measuring the documentation time of users on discharge summaries remains challenging, but will be necessary for future extrinsic evaluation of LLM\-assisted documentation\.

## Submission history

From: Nettuno Nadalini \[[view email](https://arxiv.org/show-email/3099e276/2604.19774)\] **\[v1\]**Fri, 27 Mar 2026 16:21:33 UTC \(448 KB\)

Similar Articles

Fully Open Meditron: An Auditable Pipeline for Clinical LLMs

arXiv cs.AI

Introduces Fully Open Meditron, the first fully open pipeline for building clinical LLMs, featuring a clinician-audited training corpus and reproducible framework, achieving state-of-the-art among fully open medical specialist models.

Do Benchmarks Underestimate LLM Performance? Evaluating Hallucination Detection With LLM-First Human-Adjudicated Assessment

arXiv cs.CL

This paper investigates whether standard benchmarks underestimate LLM performance by re-evaluating hallucination detection datasets using an LLM-first, human-adjudicated assessment method. The study finds that incorporating LLM reasoning into the adjudication process improves agreement and suggests that model-assisted re-evaluation yields more reliable benchmarks for ambiguity-prone tasks.

MedAction: Towards Active Multi-turn Clinical Diagnostic LLMs

arXiv cs.CL

This paper introduces MedAction, a framework for training LLMs on active, multi-turn clinical diagnosis by simulating iterative test ordering and hypothesis updates. It presents a new dataset, MedAction-32K, and demonstrates state-of-the-art performance for open-source models on medical benchmarks.

Improving health literacy and patient well-being

OpenAI Blog

Lifespan health system used GPT-4 to simplify surgical consent forms from three pages to one page at a 6th grade reading level, improving patient understanding and physician adoption. The initiative, deployed in September 2023, has received positive feedback from both patients and clinicians.