Off-the-Shelf Large Language Models for Causality Assessment of Individual Case Safety Reports: A Proof-of-Concept with COVID-19 Vaccines
Notable limitations of ChatGPT and Gemini have been identified in their use for aiding causality assessments in vaccine safety. ChatGPT performed better, with higher adherence and agreement with human experts. In the investigated scenario, both models are better suited as complementary tools to human expertise.
Authors
- Andrew Bate (ORCID: https://orcid.org/0000-0003-3151-3653)
- Anders Hviid (ORCID: https://orcid.org/0000-0002-7509-9127)
- Maurizio Sessa (ORCID: https://orcid.org/0000-0003-0874-4744)
- Andrea Abate (ORCID: https://orcid.org/0000-0002-0835-0916)
- María Antonietta Barbieri (ORCID: https://orcid.org/0000-0002-2019-4696)
- Andrea Rossi (ORCID: https://orcid.org/0009-0001-6475-2418)
- Greg Powell
- Elisa Poncato
- Simay Peker
Institutions
- Statens Serum Institut (DK)
- University of Messina (IT)
- University of Copenhagen (DK)
- University of Milan (IT)
- London School of Hygiene & Tropical Medicine (GB)
Publication Details
- Journal
- Drug Safety
- Published
- 2025-03-12
- DOI
- https://doi.org/10.1007/s40264-025-01531-y
- Citations
- 2
- Primary Topic
- Topic Modeling
- Type
- article
- Field-Weighted Citation Impact
- 4.96
Funders
- Københavns Universitet