Patient Attitudes Towards the Use of Large Language Models in Otorhinolaryngology
Overview
A survey of ORL-HNS patients revealed high familiarity with large language models (LLMs) but limited trust in their medical accuracy. Patients showed greater confidence in recommendations made by doctors compared to those made solely by LLMs.
Background
The integration of artificial intelligence (AI) and large language models (LLMs) in clinical practice is gaining traction, particularly in Otorhinolaryngology and Head and Neck Surgery (ORL-HNS). Understanding patient attitudes towards these technologies is crucial.
Data Highlights
Finding
Value
Patients familiar with AI
96% (n = 118)
Patients familiar with LLMs
78% (n = 96)
Patients using LLMs
59% (n = 72)
Confidence in LLMs' medical accuracy
Median 4 (IQR 3, 5)
Confidence in doctors' recommendations
Median 5 (IQR 5, 6)
Key Findings
96% of patients were familiar with AI.
78% of patients reported familiarity with LLMs.
59% of patients indicated they had used LLMs.
Patients rated LLMs' comprehensibility, conciseness, and coherence with a median score of 5.
Confidence in recommendations from LLMs was significantly lower than that from doctors.
Clinical Implications
Healthcare providers should be aware of patients' limited trust in LLMs for medical information.
Conclusion
While ORL-HNS patients are generally familiar with and use LLMs, their trust in these models for medical information remains limited.
These 10 metropolitan areas reported population gains and active federal shortage area designations identified in census and health care workforce data.
