Work

Health Information Seeking Behavior Using Large Language Models: A Scoping Review

Year: 2026
Type: preprint
Abstract: Rationale: Health information seeking behavior refers to the manner in which individuals seek for and use health information, including efforts to comprehend health risks and obtain ways for disease prevention and help with a prevailing diagnosis. Recently, with the advent of large language models (LLMs), health information seeking has progressed further to encompass the use of LLM-based chatbots such as ChatGPT, Microsoft Copilot and Gemini’s Google AI Overviews. With AI tools, this paradigm of traditional information seeking behavior is being revolutionized. In prior online health information seeking, the traditional search engines provide information from diverse sources, and information seekers then evaluate them. In contrast, LLM-based AI tools are not only performing the information search, but also providing curated information that is selected according to their own criteria to be the most relevant response to the seeker’s query, and offer little opportunity for individuals to pick the information. This area of research is just emerging, and the extent to which the various facets of health information seeking behavior using LLM-based tools have been examined is unclear. Study Objectives: Synthesize existing evidence on how LLMs shape, support, or transform health information seeking behavior among health consumers and caregivers. More specifically, the study aims to examine: 1) The characteristics of studies on consumers’ health information seeking behaviors using LLMs, e.g., a) Characteristics of included studies, e.g., country, methods b) Information dimension, i.e., type of information/health domain c) Source dimension, i.e., prevalent LLM tools used 2) The factors associated with health information seeking behavior using LLMs, e.g., trust, familiarity, perceived ease of use, etc. 3) The roles of LLMs in supporting health information seeking behavior, especially those that differ from the traditional online health information seeking.
Language: English
Open Access status: green
OpenAlex ID: W7161459610
Cites:
Cited by:
Related to: 10