Using ChatGPT for writing hospital inpatient discharge summaries - perspectives from an inpatient infectious diseases service.

ABSTRACT:

Background

Hospital discharge summaries are important tools for communication between healthcare professionals. They convey events that occurred during hospitalisation, as well as the subsequent follow-up plans. Artificial intelligence models can be used to summarise information succinctly from large amounts of raw data input. We explored ChatGPT's ability to generate effective discharge summaries to assist junior doctors in writing these documents.

Methods

We constructed three hypothetical scenarios of inpatient encounters, with three different outcomes: i) discharge home with follow-up with a general practitioner, ii) discharge to a stepdown facility for further physical rehabilitation, iii) transfer to a tertiary centre for more advanced care. ChatGPT was used to generate

SUBMITTER: Koh MCY 

PROVIDER: S-EPMC11809107 | biostudies-literature | 2025 Feb

REPOSITORIES: biostudies-literature