A Comparison of Ten Large Language Models and a Conventional Search Engine for Clinical Decision Support in Anesthesiology: Expert Agreement and Physician Perceptions
Main finding · Mean evaluator score (10-point Likert)Result — go to the resultF 5.89Authors' conclusion
Although LLM-generated responses differed in quality, DeepSeek R1 and Claude-Sonnet 3.5 produced answers most consistent with expert clinical judgment.
Study context
- Population
- Ten common general anesthesia questions, responses rated by 4 board-certified anesthesiologists10 questions · 4 raters
- Comparison
- Ten large language models (LLMs) vs Google search
- Outcome
- Mean evaluator score (accuracy, completeness, clinical relevance) on a 10-point Likert scale
Methodological detailsexpert-rated, 10-point Likert scale, ad hoc physician questionnaire
- Expert-rated
- 10-point Likert scale
- Ad hoc physician questionnaire
Main finding
Mean evaluator score (10-point Likert)
F
5.89
P < 0.0001
By group
DeepSeek R1
7.7
Gemini 2.0
5.2
Google Snippet Search
5.3
As reported in the abstract
Further findings
Physician Perception Questionnaire
- Limited artificial intelligence (AI) use in clinical practice; clinicians prioritized source credibility and favored traditional resources
Original publication
A Comparison of Ten Large Language Models and a Conventional Search Engine for Clinical Decision Support in Anesthesiology: Expert Agreement and Physician Perceptions. Anesth Analg. Published 29 January 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →