Diagnostic accuracy of large language models for emergency department triage: a systematic review and meta-analysis
Limei Cui, Mengting Wang, Ye Xu, Wei Wei, Xuehong Zhao · 2026
Read Diagnostic accuracy of large language models for emergency department triage: a systematic review and meta-analysis by Limei Cui, Mengting Wang, Ye Xu, Wei Wei, Xuehong Zhao (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13520277.txt.
Description
Limei Cui leads a five-author systematic review and meta-analysis in 2026 grading how accurately large language models sort emergency department patients by urgency, compared with triage nurses. Eight databases yielded 11 studies covering 3,088 real or simulated cases through March 2025.
Pooled accuracy comparisons against nurses were inconclusive, with an odds ratio of 0.65 amid very high inconsistency between studies. For spotting the sickest patients, models caught only 61 percent while correctly clearing 97 percent of others, giving moderate overall discrimination. Cui and colleagues conclude the tools might assist triage as adjuncts, yet their weak sensitivity for critical cases bars unsupervised use until prospective safety trials are done.
Catalog record
| CREATOR | Limei Cui, Mengting Wang, Ye Xu, Wei Wei, Xuehong Zhao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 47849 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |