Detecting Sociodemographic Biases in the Content and Quality of Large Language Model–Generated Nursing Care: Cross-Sectional Simulation Study
This study identified that LLMs systematically reproduce sociodemographic biases in the generation of nursing care plans. These biases appear in two forms: they shape the thematic content and influence expert-rated clinical quality. These findings reveal a substantial risk that such models may reinforce existing health inequities. To our knowledge, this is the first empirical evidence documenting these nuanced biases in nursing. The study also contributes a replicable framework for evaluating LLM-generated care plans. Finally, it underscores the critical need for robust human oversight to ensure that artificial intelligence serves as a tool for advancing equity rather than perpetuating disparities.
Authors
- Bing Xiang Yang (ORCID: https://orcid.org/0000-0002-0227-4342)
- Wei Zhai (ORCID: https://orcid.org/0009-0009-1762-9570)
- Huijing Zou (ORCID: https://orcid.org/0000-0002-5656-5364)
- Nan Bai (ORCID: https://orcid.org/0000-0002-8655-5888)
- Qing Zhao (ORCID: https://orcid.org/0000-0001-9570-9546)
- Si Chen Zhou (ORCID: https://orcid.org/0000-0001-7793-6876)
- Guanghui Fu (ORCID: https://orcid.org/0000-0002-6391-5983)
- Xinni Wei (ORCID: https://orcid.org/0000-0003-2986-012X)
- Qian Liu (ORCID: https://orcid.org/0000-0001-6833-6131)
- Jianqiang Li (ORCID: https://orcid.org/0000-0003-1995-9249)
- Luo Chun-yan (ORCID: https://orcid.org/0009-0001-8283-9015)
- Ying Yu (ORCID: https://orcid.org/0009-0002-0951-6823)
- Qing Wang (ORCID: https://orcid.org/0009-0001-8260-2501)
Institutions
- Centre National de la Recherche Scientifique (FR)
- Inserm (FR)
- Wuhan University (CN)
- Beijing University of Technology (CN)
- Sorbonne Université (FR)
- Assistance Publique – Hôpitaux de Paris (FR)
- Wuhan Donghu University (CN)
- Wuchang University of Technology (CN)
- Pitié-Salpêtrière Hospital (FR)
- Renmin Hospital of Wuhan University (CN)
- Institut du Cerveau (FR)
Publication Details
- Journal
- Journal of Medical Internet Research
- Published
- 2025-12-05
- DOI
- https://doi.org/10.2196/78132
- Citations
- 1
- Primary Topic
- Nursing education and management
- Type
- article
- Field-Weighted Citation Impact
- 5.41