Large language model-driven agents in nursing practice: A scoping review.
Xinglin Zheng, Huina Zou, Linjing Wu and 3 others
PMID 41367594WHAT IT FOUND
AI agents built on large language models are mostly prototype or simulation work in nursing, not proven patient care.
They are being tested for clinical, home, and community tasks, but outcomes need trials.
Key findings
01The review identified 25 studies, all published between 2023 and 2025, on large language model-driven agents in nursing.
02The review grouped the studies into collaborative (10 studies), augmented (9 studies), and interactive (6 studies).
03The included studies mainly tested technology rather than patient care, and the review flagged data privacy, unreliable outputs, and unclear responsibility as key barriers.
STILL TO COME
How it was doneWhat they foundWhat it means for RNs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The review mapped 25 studies, and most were technology validation or simulation rather than real-world clinical trials with patient outcomes. Some included studies did not clearly define sample sizes. The search was limited to English and Chinese papers. Most included studies came from China and the United States, so conclusions may not fit resource-constrained regions. The authors classified architectures into only collaborative, augmented, and interactive types, which may miss other technical approaches. Many studies raised ethical concerns but few proposed concrete solutions.
Declared interests
The authors declared no conflict of interest.
The easy way to misread this
Do not read this review as proof that LLM-driven agents improve nursing care. The included studies were mostly technology validation or simulation, and the review found no real-world patient outcome evidence.