Comparing large language models for antibiotic prescribing in different clinical scenarios: which performs better?
There is significant variability among LLMs in prescribing appropriate antibiotics, dosages, and treatment durations. ChatGPT-o1 outperformed other models, indicating the potential of advanced LLMs as decision-support tools in antibiotic prescribing. However, decreased accuracy in complex cases and inconsistencies among models highlight the need for careful validation before clinical utilization.
Authors
- Susan K. Seo (ORCID: https://orcid.org/0000-0002-3370-4691)
- Maria Mazzitelli (ORCID: https://orcid.org/0000-0003-0263-0703)
- Nicholas Geremia (ORCID: https://orcid.org/0000-0001-7988-5419)
- Saverio Giuseppe Parisi (ORCID: https://orcid.org/0000-0002-8865-5460)
- Anna Maria Cattelan (ORCID: https://orcid.org/0000-0003-2869-2945)
- Luigi Angelo Vaira (ORCID: https://orcid.org/0000-0002-7789-145X)
- Davide Fiore Bavaro (ORCID: https://orcid.org/0000-0002-4833-7118)
- Agnese Colpani (ORCID: https://orcid.org/0000-0002-8959-1403)
- Andrea De Vito (ORCID: https://orcid.org/0000-0002-9802-7716)
- Alberto Enrico Maraolo (ORCID: https://orcid.org/0000-0002-7218-7762)
- Giordano Madeddu (ORCID: https://orcid.org/0000-0001-6099-2273)
- Andrea Marıno (ORCID: https://orcid.org/0000-0002-5650-6911)
- Cristina Mussini (ORCID: https://orcid.org/0000-0002-6615-4198)
- Justin Laracy (ORCID: https://orcid.org/0000-0003-0710-565X)
- Michele Bartoletti (ORCID: https://orcid.org/0000-0002-1099-3283)
- Antonio Russo
- Giuseppe Nunnari
Institutions
- University of Modena and Reggio Emilia (IT)
- Memorial Sloan Kettering Cancer Center (US)
- University of Padua (IT)
- University of Sassari (IT)
- University of Catania (IT)
- Ospedale Garibaldi (IT)
- Office of Infectious Diseases (US)
- Federico II University Hospital (IT)
- University of Campania "Luigi Vanvitelli" (IT)
- University of Naples Federico II (IT)
Publication Details
- Journal
- Clinical Microbiology and Infection
- Published
- 2025-03-19
- DOI
- https://doi.org/10.1016/j.cmi.2025.03.002
- Citations
- 20
- Primary Topic
- Antibiotic Use and Resistance
- Type
- article
- Field-Weighted Citation Impact
- 25.24