跳到正文
原文
Nature 健康护理主题·· 2026-10-08精选AI 评分64

ClinicalGPT 订单时点用药安全审计基础模型:跨医疗系统回顾性评估

An audit-ready EHR foundation model for order-time medication-safety auditing across health systems

AI 导读

研究团队在 7 个医疗系统回顾性评估了 847,000 条高危用药订单,提出使用 ClinicalGPT 模型进行订单时点用药安全审计。在冻结操作点下,ClinicalGPT 配合硬约束的漏报率为每 1000 条订单 3.2 条,低于纯规则的 9.4 条和最强 LLM 比较组的 4.8 条;其虚假警报率为 33.9。

推荐理由

原文明确了 ClinicalGPT 在订单时点用药安全审计中的漏报率指标及其局限性,读者可据此评估 AI 监测工具的离线过程绩效边界。

正文 · 原文

npj Digital Medicine (2026) Cite this article

We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record. All legal disclaimers apply.

Abstract

Real-time medication-safety auditing requires current laboratory and medication histories, yet electronic health records (EHRs) may contain stale or missing decisive evidence. We present ClinicalGPT, an audit-ready foundation model using Clinical Trajectory Language to encode time, units, provenance, and observability, and workflow-aligned objectives to handle structured missingness. Across seven health systems, we retrospectively evaluated 847,000 eligible high-risk medication orders against a reference vector frozen before final scoring. The primary endpoint was the reference-standard-defined order-time failure-to-alert rate: eligible tier-i/ii reference-positive orders receiving no actionable alert at a system’s frozen operating point, per 1000 eligible orders. This offline process metric did not measure subsequent clinical actions, adverse drug events, or patient harm. ClinicalGPT with hard constraints achieved 3.2 misses per 1000 orders, versus 9.4 for rules only, 6.1 for the original generic LLM comparator, and 4.8 for the strongest observed LLM comparator (RAG + LLM + rules). Its false-alert rate was 33.9 (95% CI, 31.6–36.3), versus 38.7 for rules only. A separate single-site silent prospective assessment evaluated workflow measures without altering care. Matched ablations supported explicit observability modeling; independent de novo labeling of an enriched knowledge-base–independent subset preserved sensitivity ordering but narrowed the ClinicalGPT–RAG gap. These findings support comparative order-time auditing performance but do not establish clinical effectiveness or reduced patient harm.

Acknowledgements

No funding was received for this research. We thank Jiayue Zou for contributions to the additional experiments conducted during manuscriptrevision. The authors used AI-assisted tools for language editing and LaTeX formatting; all content was reviewed and approved by the authors, who take responsibility for its accuracy and integrity.

Author information

Author notes

  1. These authors contributed equally: Shuang Cao, Rui Li.

Authors and Affiliations

  1. Hill Research, Princeton, NJ, USA

    Shuang Cao, Rui Li, Ruihua Liu & Alexandre Duprey

Authors

  1. Shuang Cao
  2. Rui Li
  3. Ruihua Liu
  4. Alexandre Duprey

Corresponding author

Correspondence to Rui Li.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Cao, S., Li, R., Liu, R. et al. An audit-ready EHR foundation model for order-time medication-safety auditing across health systems. npj Digit. Med. (2026). https://doi.org/10.1038/s41746-026-03243-8

Download citation

  • Received: 28 December 2025

  • Accepted: 31 August 2026

  • Published: 08 October 2026

  • DOI: https://doi.org/10.1038/s41746-026-03243-8

来源:Nature 健康护理主题 · nature.com