Using secure artificial intelligence agents integrated within the electronic medical record for the evaluation of blood culture appropriateness—Northern California, 2025

2025/11/11 by Guillermo Rodriguez‐Nava, Timothy Keyes, Nerissa Ambers +5 · 1 voice · 1 citation
Biochemistry, Genetics and Molecular Biology · Health Professions · Medicine · #Artificial Intelligence in Healthcare and Education #Bacterial Identification and Susceptibility Testing #Electronic Health Records Systems

paper · pdf · doi:10.1017/ice.2025.10349

openalex created_date 2025/11/11 · openalex publication_date 2025/11/11 · openalex updated_date 2026/05/21

Abstract

We evaluated large language model (LLM)-based agents integrated with the electronic medical record to assess blood culture appropriateness. While sensitivity was high, specificity remained low. Performance was shaped by prompt phrasing, sycophantic behavior, and semantic triggers, reflecting both the potential and limitations of LLMs in real-world clinical decision support.

Citations

• AI Supported Degradation of the Self Concept: A Theoretical Framework Grounded in Established Cognitive and Computational Mechanisms
• Performance of a large language model for identifying central line-associated bloodstream infections (CLABSI) using real clinical notes
• Unleashing the potential of prompt engineering for large language models

Cited by

• A protocol for retrospective evaluation of potential patient-level harms of an AI tool for blood culture stewardship

Discussions

• New in ICHE from @grod11.bsky.social et al: Using secure AI agents integrated within the EMR for the evaluation of blood culture appropriateness LLM agents showed: ▶️ High sensitivity (77.4%) ▶️ Low [bsky, 3 points, 1 comments]

Related

• A Multi-Phase Analysis of Blood Culture Stewardship: Machine Learning Prediction, Expert Recommendation Assessment, and LLM Automation
• Large Language Models Illuminate a Progressive Pathway to Artificial Healthcare Assistant: A Review
• A protocol for retrospective evaluation of potential patient-level harms of an AI tool for blood culture stewardship
• Large Language Models as Agents in the Clinic
• A Field Guide to Deploying AI Agents in Clinical Practice