Skip to content
Review Open access

Bias and equity in large language model-based applications for health care: a scoping review.

Nov 2026 · The Lancet Digital Health · pp. 101112 · 0 citations · 243 references
Medicine

Abstract

The integration of large language models (LLMs) into health care presents both opportunities for innovation and risks of perpetuating existing biases and health disparities. Although these concerns are widely discussed, a systematic understanding of the research landscape regarding bias and equity in LLM-based applications for health care remains limited, and few discussions have translated into actionable mitigation strategies. We conducted a Scoping Review to systematically map this landscape and identify key trends and gaps. We searched PubMed, Embase, Web of Science, and Scopus for English-language articles published from Jan 1, 2018, to March 10, 2026. Our analysis of the 201 included studies revealed a strong focus on specific medical contexts and demographic factors and a predominant reliance on proprietary models that constrain transparency in bias evaluation. A disconnect exists between bias detection and mitigation, and the potential of LLMs to promote health equity remains underexplored. To realise a more coherent integration of bias mitigation and health equity objectives in health-care LLM development, the field needs large-scale, endpoint-driven studies grounded in real-world clinical workflows. Furthermore, we advocate for standardised multidimensional bias evaluation frameworks developed with support from professional societies and lifecycle bias governance across both proprietary and open-weight models to guide the safe and equitable integration of LLMs into health care globally.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.