Attitudes Toward Large Language Models in Health Care and Preferences for Their Adoption and Oversight Among Health Care Professionals: Cross-Sectional Survey - Report - MDSpire
Clinical Report: Health Care Professionals' Perspectives on LLM Integration
Background
The rapid adoption of large language models in healthcare has significant limitations, such as inaccuracies and ethical concerns, necessitating careful consideration of their integration into clinical practice.
Data Highlights
No numerical or trial data is provided in the source material.
Key Findings
LLM adoption among physicians increased from 38% in 2023 to 66% in 2024, as reported by the American Medical Association.
LLMs can produce false or hallucinated information, raising concerns in high-stakes clinical settings, as noted in various studies.
Algorithmic bias in LLMs may exacerbate existing disparities in healthcare access and outcomes, according to recent research.
Regulatory frameworks for LLMs in healthcare are fragmented and lack standardized predeployment validation requirements, as highlighted in the literature.
Ethical issues surrounding LLMs include biases in training datasets and limited transparency in model development, as discussed in multiple sources.
Clinical Implications
Healthcare professionals must be aware of the limitations of LLMs, particularly regarding accuracy and bias.
Conclusion
The integration of large language models in healthcare presents challenges that require careful management.
by Arya Rao, Chinemerem Nwokemodo-Ihejirika, John W R Kincaid, Marharyta Krylova, Kaiz P Esmail, Dan Nguyen, Christian Rivera, Erica Koranteng, Marc D Succi
