This systematic review investigates bias in Large Language Models (LLMs) across healthcare applications, revealing significant implications for patient care and health equity.

• Analyzes the prevalence and manifestations of bias in clinical LLM applications
• Identifies sources of bias that could compromise patient safety
• Explores how biases may exacerbate existing health inequities
• Provides a foundation for developing more equitable AI systems in healthcare

This research is critical for medical professionals and AI developers as biased healthcare AI could perpetuate discrimination, compromise patient safety, and worsen disparities in care quality across diverse populations.

Bias in Large Language Models Across Clinical Applications: A Systematic Review