Decision models can offload structured clinical tasks from large language models
- 1Department of Family Medicine, Clalit Health Services
Abstract
Decision models like Jev read text and return probabilities, categories, or ordinal levels, with confidence scores for the latter two, at a fraction of large language models (LLMs) cost. We explored how Jev could be integrated into clinical workflows for all these output types: mortality risk estimation, categorical Choosing-Wisely decisions for patient questions, and ordinal colon cancer staging. With standard LLM-style inputs, Jev trailed five flagship LLMs: mortality AUROC was 0.822 versus 0.837-0.846; Choosing-Wisely accuracy, 87.7% versus 98.4-100%; and staging accuracy, 45.4% versus 94.2-100%. Developer-recommended input alterations improved staging accuracy to 75.8% with medical context, 82.1% with task decomposition, and 85.8% with both. None of Jevs staging errors had full confidence. Using Jev for fully confident staging decisions and LLMs for the rest was non-inferior to each LLM alone and cut flagship LLM costs by 33-44%. Prompted as designed, decision models may efficiently share structured clinical tasks with LLMs.
Related
- Extracting smoking history from clinical notes for lung cancer screening decision support: comparing a structured-judgment model with general-purpose large language models
- Large language model-assisted CAD-RADS classification: structured report extraction, accuracy, and clinical decision-making risk of ChatGPT-4o/5 and DeepSeek-R1
- Influence of large language models on pediatricians’ clinical reasoning and decision-making process
- Reliability, Quality, and Readability of Freely Available Large Language Models for Clinical Decision Support in Perforated Peptic Ulcer: A Blinded, Multidimensional Study
- The Influence of Large Language Model-Generated Artificial Intelligence Risk Estimates on Clinical Decision-Making in Breast Imaging: A Simulation-Based Study
BibTeX
@misc{ariel2026decision,
title = {{Decision models can offload structured clinical tasks from large language models}},
author = {D. Ariel and G. Laban and S. Endelman and R. Balicer and N. Dagan and S. Cohen},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.09.26365130},
url = {https://www.medrxiv.org/content/10.64898/2026.10.09.26365130v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Decision models can offload structured clinical tasks from large language models
AU - D. Ariel
AU - G. Laban
AU - S. Endelman
AU - R. Balicer
AU - N. Dagan
AU - S. Cohen
PY - 2026
DA - 2026/10/09
PB - medRxiv
DO - 10.64898/2026.10.09.26365130
UR - https://www.medrxiv.org/content/10.64898/2026.10.09.26365130v1
AB - Decision models like Jev read text and return probabilities, categories, or ordinal levels, with confidence scores for the latter two, at a fraction of large language models (LLMs) cost. We explored how Jev could be integrated into clinical workflows for all these output types: mortality risk estimation, categorical Choosing-Wisely decisions for patient questions, and ordinal colon cancer staging. With standard LLM-style inputs, Jev trailed five flagship LLMs: mortality AUROC was 0.822 versus 0.837-0.846; Choosing-Wisely accuracy, 87.7% versus 98.4-100%; and staging accuracy, 45.4% versus 94.2-100%. Developer-recommended input alterations improved staging accuracy to 75.8% with medical context, 82.1% with task decomposition, and 85.8% with both. None of Jevs staging errors had full confidence. Using Jev for fully confident staging decisions and LLMs for the rest was non-inferior to each LLM alone and cut flagship LLM costs by 33-44%. Prompted as designed, decision models may efficiently share structured clinical tasks with LLMs.
ER -