The abstract is read at the publisher; the summary is JClub's.
Across 201 health-care LLM studies (Jan 1, 2018–Mar 10, 2026), research heavily emphasizes specific medical contexts and demographics while relying mainly on proprietary models that limit bias transparency.
- Why it matters: Bias and equity risks in health-care LLMs are widely discussed, but the research landscape is not well mapped and mitigation guidance is not consistently actionable.
- What they did: A scoping review searched PubMed, Embase, Web of Science, and Scopus for English-language articles from Jan 1, 2018, to Mar 10, 2026, then analyzed trends and gaps.
- The result: The review finds a disconnect between bias detection and mitigation and that health-equity potential is underexplored, implying a need for large, endpoint-driven studies in real clinical workflows and standardized bias evaluation with lifecycle governance.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.