Benchmarking the Confidence of Large Language Models in Answering Clinical Questions: Cross-Sectional Evaluation Study
JMIR Medical Informatics · 16.05.2025 · 13 · e66917-e66917
DOI: 10.2196/66917 ↗Kurz gefasst
Querschnittstudie in der Zeitschrift JMIR Medical Informatics, online erschienen am 16.05.2025. Verfasst von Mahmud Omar und 4 weiteren Autor:innen. Beteiligte Einrichtung: Division of Data-Driven and Digital Medicine (D3M), Department of Medicine, Icahn School of Medicine at Mount Sinai, Gustave L. Levy Place New York, …. Hauptthemen laut MEDLINE: Benchmarking. Die Arbeit wurde bisher 64-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Background
The capabilities of large language models (LLMs) to self-assess their own confidence in answering questions within the biomedical realm remain underexplored.
Objective
This study evaluates the confidence levels of 12 LLMs across 5 medical specialties to assess LLMs’ ability to accurately judge their own responses.
Methods
We used 1965 multiple-choice questions that assessed clinical knowledge in the following areas: internal medicine, obstetrics and gynecology, psychiatry, pediatrics, and general surgery. Models were prompted to provide answers and to also provide their confidence for the correct answers (score: range 0%‐100%). We calculated the correlation between each model’s mean confidence score for correct answers and the overall accuracy of each model across all questions. The confidence scores for correct and incorrect answers were also analyzed to determine the mean difference in confidence, using 2-sample, 2-tailed t tests.
Results
The correlation between the mean confidence scores for correct answers and model accuracy was inverse and statistically significant ( r =−0.40; P =.001), indicating that worse-performing models exhibited paradoxically higher confidence. For instance, a top-performing model—GPT-4o—had a mean accuracy of 74% (SD 9.4%), with a mean confidence of 63% (SD 8.3%), whereas a low-performing model—Qwen2-7B—showed a mean accuracy of 46% (SD 10.5%) but a mean confidence of 76% (SD 11.7%). The mean difference in confidence between correct and incorrect responses was low for all models, ranging from 0.6% to 5.4%, with GPT-4o having the highest mean difference (5.4%, SD 2.3%; P =.003).
Conclusions
Better-performing LLMs show more aligned overall confidence levels. However, even the most accurate models still show minimal variation in confidence between right and wrong answers. This may limit their safe use in clinical settings. Addressing overconfidence could involve refining calibration methods, performing domain-specific fine-tuning, and involving human oversight when decisions carry high risks. Further research is needed to improve these strategies before broader clinical adoption of LLMs.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Division of Data-Driven and Digital Medicine (D3M), Department of Medicine, Icahn School of Medicine at Mount Sinai, Gustave L. Levy Place New York, New York, NY, 10029, United States, 1 212 241 6500.
- Ophthalmology Department, Hadassah Medical Center, Jerusalem, Israel.
ORCID-Profile: Mahmud Omar, Reem Agbareia, Benjamin S Glicksberg, Girish N Nadkarni, Eyal Klang
Zitieren
APA
Omar, M., Agbareia, R., Glicksberg, B. S., Nadkarni, G. N., & Klang, E. (2025). Benchmarking the Confidence of Large Language Models in Answering Clinical Questions: Cross-Sectional Evaluation Study. JMIR Medical Informatics, 13, e66917-e66917. https://doi.org/10.2196/66917
Vancouver
Omar M, Agbareia R, Glicksberg BS, Nadkarni GN, Klang E. Benchmarking the Confidence of Large Language Models in Answering Clinical Questions: Cross-Sectional Evaluation Study. JMIR Medical Informatics. 2025;13:e66917-e66917. doi:10.2196/66917
BibTeX
@article{Omar2025,
author = {Omar, Mahmud and Agbareia, Reem and Glicksberg, Benjamin S and Nadkarni, Girish N and Klang, Eyal},
title = {Benchmarking the Confidence of Large Language Models in Answering Clinical Questions: Cross-Sectional Evaluation Study},
journal = {JMIR Medical Informatics},
year = {2025},
volume = {13},
pages = {e66917--e66917},
doi = {10.2196/66917}
}
EUVIMED