Large language models (LLMs) are increasingly being explored for patient education and clinical decision support in otolaryngology. However, their reliability varies across clinical tasks, and their performance in otology and neurotology, where decision-making often requires integration of complex audiological, vestibular, and imaging data, remains unclear.
✦ The floor
Discussion
Signed responses from readers of the wire.
No actionable practice change yet — LLMs show promise for otology/neurotology support but systematic evidence on accuracy and safety is still emerging; clinicians should treat AI-generated guidance as supplementary and unverified.
As AI chatbots enter clinical workflows, audiology and otology need rigorous evidence on their reliability before widespread adoption in patient education or decision support.
- 01Systematic review and meta-analysis of LLM performance in otology and neurotology contexts.
- 02Assessed two main use cases: patient education and clinical decision support.
- 03Evidence base for LLM clinical utility in this specialty is still early-stage.
- 04Reliability and accuracy of LLMs vary across tasks and conditions studied.
- 05Findings will inform whether AI chatbots can safely supplement specialist care.
LLMs have clinical utility for patient education and decision support in otology and neurotology.
studypartially supported- PMID
- 42931836
- DOI
- 10.1007/s00405-026-10725-8.
- Journal
- European Archives of Oto-Rhino-Laryngology
- Publication type
- meta_analysis
- Evidence level
- 1a
- Population
- Studies evaluating LLM performance in otology and neurotology clinical contexts
- Intervention
- Large language models (LLMs) for patient education and clinical decision support
Primary outcomes
Clinical utility of LLMs in otology and neurotology; Reliability and accuracy of LLM-generated content