Potential applications and implications of large language models in primary care

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Digital Health

Background:

• Generative artificial intelligence (AI) chatbots, such as ChatGPT and Bard, powered by large language models (LLMs), are gaining attention for healthcare applications.
• LLMs show potential in clinical decision-making, administrative task streamlining, and patient empowerment.
• Existing research often focuses on broad medical applications, with limited exploration of primary care.

Purpose of the Study:

• To consolidate current literature on LLMs in primary care.
• To summarize the potential benefits, risks, and challenges of LLM integration in primary care settings.
• To provide insights for primary care clinicians on adopting AI technologies.

Main Methods:

• Literature review of current and pertinent research on LLMs in medicine.
• Analysis of potential applications, benefits, and limitations of LLMs in primary care.
• Identification of ethical considerations and practical insights for clinical adoption.

Main Results:

• LLMs offer significant potential benefits for primary care, including enhanced decision support and administrative efficiency.
• Key risks and challenges include privacy concerns, data bias, and ensuring the accuracy and reliability of AI-generated information.
• The study highlights the novelty of examining LLMs specifically within the primary care context.

Conclusions:

• LLMs present a transformative potential for primary care, but careful consideration of ethical and practical challenges is crucial.
• Primary care clinicians need guidance on evaluating and integrating these AI tools responsibly.
• Further research and policy development are needed to navigate the complexities of AI in primary care.