Consumer and Patient Health Information Seeking With Generative AI Tools: Scoping Review of Facilitators and Barriers

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background Generative AI (GenAI) tools powered by large language models (LLMs) are increasingly used by the public to seek health information. Unlike traditional web search, these systems generate conversational responses that may alter how users assess credibility, manage uncertainty, verify information, and decide whether to consult clinicians. As GenAI becomes more embedded in everyday health information practices, a clearer synthesis of the emerging empirical evidence is needed. Objective This scoping review mapped and synthesized empirical research on consumer and patient health information seeking using GenAI and LLM tools, with a focus on study contexts, outcome constructs, and the facilitators and barriers shaping use, reliance, and verification. Methods The review adhered to Joanna Briggs Institute guidance for scoping reviews and reported using PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews), with search reporting additionally guided by PRISMA-S (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Search Extension). We searched PubMed, Scopus, PsycINFO, Web of Science, IEEE Xplore, ACM Digital Library, Google Scholar, ERIC, EBSCO, and ProQuest for English-language studies published 2022 onward. The final updated search was conducted on January 8, 2026. Eligible studies were empirical quantitative, qualitative, or mixed methods studies examining health information seeking mediated by GenAI and LLM systems, wherein an LLM served as the interface or source for obtaining health information. Data were charted using a structured extraction form capturing study characteristics, populations, health contexts, GenAI tool types, outcomes, and factors shaping use. Results The review included 27 studies. GenAI was used for symptom appraisal, condition understanding, treatment options, and care navigation. Reported facilitators included convenience and clarity, particularly efficiency and access (n=8, 29.6%), comprehensibility and presentation quality (n=11, 40.7%), personalization and specificity (n=5, 18.5%), and affective or interpersonal comfort (n=5, 18.5%). Reported barriers were dominated by credibility and trust concerns (n=13, 48.1%), particularly when accuracy cues or citations were missing or difficult to interpret. Additional barriers included perceived unsuitability for complex, urgent, or emotionally charged situations (n=5, 18.5%); privacy or data security concerns (n=4, 14.8%); limited prompting skills (n=2, 7.4%); and modality or interaction constraints that hindered credibility assessment and information comparison (n=5, 18.5%). Six (22.2%) studies reported literacy-related capability was, and 5 (18.5%) reported verification-supporting features, such as visible sourcing, transcripts, and save, revisit, or share functions. Conclusions This review is innovative in focusing on health information seeking as a user practice rather than on technical performance or clinical implementation alone. Unlike prior reviews, it maps how the emerging literature conceptualizes use, trust, reliance, and verification. It contributes a structured synthesis of the main facilitators, barriers, and verification-related features reported on GenAI-mediated health information seeking. In practice, the findings suggest that safer use may depend on not only model quality but also users’ ability to interpret, verify, and act on AI-generated responses.