Codesigning a Nurse-Led, Large Language Model-Empowered Agent to Increase Hepatitis B Screening and Vaccination for...

Area of Science:

• Public Health
• Artificial Intelligence in Healthcare
• Nursing Informatics

Background:

• Hepatitis B screening and vaccination rates are suboptimal in inclusion health populations.
• Addressing screening and vaccination disparities requires innovative, targeted interventions.

Purpose of the Study:

• To codesign and test a nurse-led agent powered by large language models (LLMs) to enhance hepatitis B screening and vaccination.
• To improve health equity and preventive care delivery for inclusion health populations.

Main Methods:

• Utilized a double diamond model-guided codesign methodology across four phases: Discover, Define, Develop, and Deliver.
• Conducted systematic reviews to identify intervention targets and effective components, including LLM adaptation.
• Developed an LLM agent for human-agent interactive consultation, intelligent case management, and care coordination, involving diverse stakeholders.

Main Results:

• The developed agent aims to facilitate nurse-led management of hepatitis B screening, vaccination, and care linkage.
• Mixed-methods evaluation will assess model response, user experience, and technical feasibility.
• Anticipated improvements in hepatitis B screening and vaccination rates within the target population.

Conclusions:

• The intervention is expected to enhance hepatitis B diagnosis, immunity, and care linkage for inclusion health populations.
• Establishes a codesign framework for nursing-specific LLMs, expanding nursing's role in preventive health equity.