Contextual Conversational Agent to Address Vaccine Hesitancy: Protocol for a Design-Based Research Study

Area of Science:

• Health communication
• Human-computer interaction
• Artificial intelligence in healthcare

Background:

• Misinformation regarding COVID-19 and vaccines is widespread.
• Health organizations utilize conversational agents (CAs) to combat misinformation.
• Effective CA design is crucial for successful health communication strategies.

Purpose of the Study:

• To describe the application of design-based research (DBR) for contextual CA design.
• To address vaccine hesitancy through tailored CA development.
• To examine DBR's impact on CA accuracy and performance.

Main Methods:

• Utilized a design-based research (DBR) methodology with iterative cycles of analysis, design, and evaluation.
• Employed the System Usability Scale for user perception assessment.
• Adopted the PARADISE (PARAdigm for Dialogue System Evaluation) method for performance calculation of the text-based CA.

Main Results:

• Completed the design and evaluation phases of the first DBR cycle.
• Currently refining the natural-language understanding model for pilot intervention.
• Planning three DBR cycles with expected results from the first cycle by end of 2022.

Conclusions:

• Conversational agents (CAs) offer an innovative channel for health information dissemination.
• Careful design is essential for CAs aiming for behavioral change or health education.
• DBR provides a framework for understanding and optimizing CA development for health communication.