Self-Diagnosis through AI-enabled Chatbot-based Symptom Checkers: User Experiences and Design Considerations

Area of Science:

• Health Informatics
• Artificial Intelligence in Healthcare
• Human-Computer Interaction

Background:

• Growing interest in AI-enabled chatbot-based symptom checker (CSC) apps for healthcare.
• CSC apps utilize AI for preliminary diagnoses and self-triaging through conversational interfaces.
• Limited research exists on the functionalities and user experiences of current CSC apps.

Purpose of the Study:

• To investigate the functionalities and user experiences of existing AI-enabled CSC apps.
• To identify limitations in current CSC app features and conversational design.
• To provide implications for future CSC app development.

Main Methods:

• A comprehensive feature review of existing CSC apps.
• Analysis of user reviews from app stores and online platforms.
• Conducting an interview study with CSC app users.

Main Results:

• Existing CSC apps lack features to fully support the diagnostic process for an offline medical visit.
• Users reported deficiencies in comprehensive medical history taking, flexible symptom input, and understandable questioning.
• Identified a lack of support for diverse diseases and user demographics within current CSC apps.

Conclusions:

• Current CSC apps require enhancement to cover the full diagnostic journey.
• Future CSC app development should prioritize robust medical history integration and adaptable conversational elements.
• Addressing user-perceived limitations is crucial for improving the efficacy and adoption of CSC technology.