Health Information Seeking From an Intelligent Web-Based Symptom Checker: Cross-sectional Questionnaire Study

Area of Science:

• Digital Health
• Health Informatics
• Artificial Intelligence in Healthcare

Background:

• Increasing demand for personalized online health information.
• Symptom checkers offer potential diagnoses but their use and effects are understudied.
• Need to understand user demographics, motivations, and outcomes of AI-powered symptom checkers.

Purpose of the Study:

• Identify users of a web-based AI symptom checker (Buoy).
• Analyze user experience, information quality, and intended actions.
• Determine predictors of future use and satisfaction across diverse groups.

Main Methods:

• Cross-sectional survey of 2437 users post-symptom checker visit.
• Assessed comprehensibility, confidence, usefulness, anxiety, empowerment, and future use intention.
• Used ANOVA, Wilcoxon rank sum test, and multilevel logistic regression for analysis.

Main Results:

• Users were predominantly well-educated, White, and female.
• Pain, gynecological issues, and masses were common reasons for use.
• Users reported high confidence, usefulness, and reduced anxiety; minority users showed higher confidence and intention to consult providers.

Conclusions:

• Web-based symptom checkers show potential for empowering users and reducing health anxiety.
• The tool may address unmet needs for women and Black/Latino adults.
• Persistent second-level digital divide effects were observed in user base analysis.