Locally Deployed Large Language Models for AI-Assisted Outpatient Prescription Review: Crossover Study
JMIR Medical Informatics · 2026
DOI: 10.2196/97520 ↗Kurz gefasst
Fachartikel in der Zeitschrift JMIR Medical Informatics, 2026. Verfasst von Zhengyue Liu und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Clinical Pharmacy, Shenyang pharmaceutical university, Shenyang, China. Hauptthemen laut MEDLINE: Large Language Models, Artificial Intelligence, Outpatients, Drug Prescriptions. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 17 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Background
Pharmacists’ prescription review is key to medication safety, but rising numbers of outpatient prescriptions and expanding formularies leave less time per case, increasing error risk. Large language models (LLMs) show promise, yet two barriers hinder routine use: first, most systems rely on cloud-based commercial models, risking data breaches by transmitting protected patient information externally, and second, retrieval-augmented generation (RAG) pipelines used to reduce hallucinations depend on complex text vectorization and vector databases, which hospitals with limited IT resources struggle to build and maintain.
Objective
This study aimed to evaluate the feasibility and usefulness of a locally deployed, knowledge-augmented LLM as a decision-support tool for pharmacist-led outpatient prescription review.
Methods
The open-source Qwen3-14B model was deployed on a hospital intranet server using the Ollama framework. A structured knowledge base derived from drug package inserts served as the principal reference and supported lightweight knowledge augmentation through exact-match injection. A 2-period crossover design was used: 2 pharmacists independently reviewed 213 outpatient prescriptions under both unaided and AI-assisted conditions, yielding paired unaided and collaborative results for each prescription. Plain AI review and knowledge-augmented AI review were also run on the same 213-prescription test set to quantify hallucination rates, and stand-alone AI performance was assessed against the reference standard. The reference standard was established by independent consensus between two supervising pharmacists, with disagreements adjudicated by a deputy chief pharmacist. Accuracy, sensitivity, and specificity were compared between conditions using paired McNemar tests, and review time was compared using the Wilcoxon signed-rank test.
Results
Overall review accuracy was 97.2% (207/213, 95% CI 94.0%-98.7%) in the human-AI collaborative condition vs 82.6% (176/213, 95% CI 77.0%-87.1%) in the pharmacist-alone condition (P
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Clinical Pharmacy, Shenyang pharmaceutical university, Shenyang, China.
- Department of Pharmacy, Shanghai Changhai Hospital, The First Affiliated Hospital of Navy Medical University, Shanghai, China.
ORCID-Profile: Zhengyue Liu, Yi Ding, Jingxia Chen, Ziqiang Yan, Xuxi Cheng, Wei Zhou, Peng Fu, Zhuo Wang
Zitieren
APA
Liu, Z., Ding, Y., Chen, J., Yan, Z., Cheng, X., Zhou, W., Fu, P., & Wang, Z. (2026). Locally Deployed Large Language Models for AI-Assisted Outpatient Prescription Review: Crossover Study. JMIR Medical Informatics. https://doi.org/10.2196/97520
Vancouver
Liu Z, Ding Y, Chen J, Yan Z, Cheng X, Zhou W, et al. Locally Deployed Large Language Models for AI-Assisted Outpatient Prescription Review: Crossover Study. JMIR Medical Informatics. 2026. doi:10.2196/97520
BibTeX
@article{Liu2026,
author = {Liu, Zhengyue and Ding, Yi and Chen, Jingxia and Yan, Ziqiang and Cheng, Xuxi and Zhou, Wei and Fu, Peng and Wang, Zhuo},
title = {Locally Deployed Large Language Models for AI-Assisted Outpatient Prescription Review: Crossover Study},
journal = {JMIR Medical Informatics},
year = {2026},
doi = {10.2196/97520}
}
EUVIMED