ArticleNature medicine2025
Sociodemographic biases in medical decision making by large language models.
Article in Nature medicine, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 92 papers, 2 of them syntheses that pooled it.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
92 citing papers in PubMed, 2 syntheses or guidelines pooled it.
- Large language models in real-world clinical workflows: a systematic review of applications and implementation.Frontiers in digital health · 2025Pooled it
- Emerging applications of NLP and large language models in gastroenterology and hepatology: a systematic review.Frontiers in medicine · 2024Pooled it
- Safety-Filter Fallback on Consumer Health Questions During the Initial Claude Fable 5 Deployment: Observational Study.JMIR AI · 2026Article
- On-premise medical AI agents for reliable clinical decision-making.Nature medicine · 2026Article
- Large language model-based triage to identify antiretroviral therapy adherence barriers and risk levels in patient messages.JAMIA open · 2026Article
- Assessing and mitigating demographic bias in large language models for diagnostic radiology.Japanese journal of radiology · 2026Article
- Cultural bias in large language models' ability to follow neuroradiology guidelines.European radiology · 2026Article
- MedNLI-Pain benchmarks a minimally necessary requirement for large language model alignment.JAMIA open · 2026Article
- Sex and gender bias in large language models: an old problem at a new scale.BMJ health & care informatics · 2026Article
- LitBench: A Benchmark for Retrieval-Grounded Multi-Paper Evidence Synthesis in Epilepsy.medRxiv : the preprint server for health sciences · 2026Article
- Process-Oriented, Behaviorally Anchored Assessment of Clinical Reasoning in Large Language Models and the Effect of Extended Thinking: Protocol for a Prospective, Multigroup, Comparative Study.JMIR research protocols · 2026Article
- [Application effect of LLMs as decision-support tools in emergency burn treatment].Zhonghua shao shang yu chuang mian xiu fu za zhi · 2026Article
- Data distribution impacts the performance and generalisability of contrastive learning-based foundation models of electrocardiograms.PLOS digital health · 2026Article
- DFGLM-TCM: an integrated knowledge-and experience-driven large language model system for Traditional Chinese Medicine practice.Chinese medicine · 2026Article
- Social Status and Clinical Resource Allocation by a Large Language Model: An Evaluation of 30,618 Decisions.Journal of personalized medicine · 2026Article
- Effect of Fitzpatrick Skin Type Prompting on Diagnostic Accuracy in Multimodal Large Language Models: A Within-Image Experimental Study.Bioengineering (Basel, Switzerland) · 2026Article
- Large language models in spine care and research.European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society · 2026Review
- Large Language Models for Mental Health Prediction: Scoping Review of Bias and Clinical Utility Documentation in 2019-2024.JMIR AI · 2026Review
- AI Integration in Undergraduate Medical Education: Qualitative Study of Faculty Perspectives in the United Arab Emirates.JMIR medical education · 2026Article
- Review
32 more citing papers are in PubMed but not listed here.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
12 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
Large language models (LLMs) show promise in healthcare, but concerns remain that they may produce medically unjustified clinical care recommendations reflecting the influence of patients' sociodemographic characteristics. We evaluated nine LLMs, analyzing over 1.7 million model-generated outputs from 1,000 emergency department cases (500 real and 500 synthetic). Each case was presented in 32 variations (31 sociodemographic groups plus a control) while holding clinical details constant. Compared to both a physician-derived baseline and each model's own control case without sociodemographic identifiers, cases labeled as Black or unhoused or identifying as LGBTQIA+ were more frequently directed toward urgent care, invasive interventions or mental health evaluations. For example, certain cases labeled as being from LGBTQIA+ subgroups were recommended mental health assessments approximately six to seven times more often than clinically indicated. Similarly, cases labeled as having high-income status received significantly more recommendations (P < 0.001) for advanced imaging tests such as computed tomography and magnetic resonance imaging, while low- and middle-income-labeled cases were often limited to basic or no further testing. After applying multiple-hypothesis corrections, these key differences persisted. Their magnitude was not supported by clinical reasoning or guidelines, suggesting that they may reflect model-driven bias, which could eventually lead to health disparities rather than acceptable clinical variation. Our findings, observed in both proprietary and open-source models, underscore the need for robust bias evaluation and mitigation strategies to ensure that LLM-driven medical advice remains equitable and patient centered.
Indexed as
Identifiers
40195448What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.