药品检查缺陷关键词风险评估研究

doi:10.19803/j.1672-8629.20260062

摘要/Abstract

摘要： 目的生成药品检查缺陷关键词库,分析关键词、关键词组合在不同情境下缺陷归类情况及风险等级。方法运用大语言模型,提取国家药品监督管理局食品药品审核查验中心药品监管检查报告中缺陷数据的关键词,通过遍历、VBA函数高度自动化等方式,计算缺陷关键词、关键词组合在其所属缺陷分类中的占比。结果构建药品检查缺陷关键词库以及检查缺陷关键词、关键词组合风险等级表,其中包含缺陷关键词、关键词组合、其所属缺陷分类情况、风险等级,并根据风险由高到低进行排序。结论采用人工智能技术与传统的药品现场检查相融合的方式,利用大语言模型工具节约大量数据归纳整理所需的人力及时间,实现药品检查工作从依赖个人经验向数据驱动的转变。

关键词: 药品检查, 检查缺陷, 缺陷, 缺陷分类, 缺陷风险等级, 药品安全

Abstract: Objective To generate a keyword library for drug inspection defects and analyze classifications and risk levels of the defects based on the keywords and their combinations in different settings. Methods The keywords were retrieved from reports about GMP inspection delivered by the Center for Food and Drug Inspection of NMPA using a large language model. Highly automated methods, including traversal algorithms and VBA functions, were employed to calculate the proportions of defect keywords and keyword groups within their respective classification. Results A detailed keyword database for GMP inspection defects was established, along with a risk level table involving defect keywords, keyword groups, their classifications, and risk levels ranked from the highest to the lowest. Conclusion This research has established a model that integrates artificial intelligence technologies with traditional on-site drug inspections, which can significantly reduce the labor and time required for data compilation and analysis. It pioneers a novel paradigm of intelligent regulatory oversight based on large language models, facilitating a shift from experience-based inspections to data-driven decision-making.

Key words: Drug Inspection, Inspection Defects, Defect, Defects Classifications, Defects Risk Levels, Drug Safety

中图分类号:

R954
R994.11

李聪慧, 高振宇, 张京梅. 药品检查缺陷关键词风险评估研究[J]. 中国药物警戒, 2026, 23(6): 667-671.

LI Conghui, GAO Zhenyu, ZHANG Jingmei. Relationships between Risk Levels and Categories of Keywords for Drug Inspection Defects[J]. Chinese Journal of Pharmacovigilance, 2026, 23(6): 667-671.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://zgywjj.magtechjournal.com/CN/10.19803/j.1672-8629.20260062

https://zgywjj.magtechjournal.com/CN/Y2026/V23/I6/667

参考文献

[1] NHC. Good Manufacturing Practice of Medical Products[EB/OL]. (2011-02-12)[2026-05-07]. https://www.nhc.gov.cn/wjw/bmgz/201102/62dcfa1956c347759674547dcaead345.shtml.
[2] LIU S.Research on Countermeasures for Changes in GMP Regulatory Mode of API Enterprises-Based on the Implementation of the New Drug Administration Law[J]. China Market(中国市场), 2021(24): 69-70.
[3] FAN XD.Overall Concept of the Information Construction of Drug Authentication[J]. China Pharmacy(中国药房), 2009, 20(25): 1925-1927.
[4] General Office of the State Council. Opinions on Establishing a Team of Professional and Specialized Drug Inspectors[EB/OL]. (2019-07-18)[2026-05-07]. https://www.gov.cn/zhengce/content/2019-07/18/content_5411172.htm.
[5] HAO LJ.The System for Professional and Specialized Drug Inspector Teams Is Becoming Increasingly Well-Established[J]. China Quality Certification(质量与认证), 2019(9): 26-27.
[6] CHEN F, WU XR.Practices and Reflections on the Application of Large Language Models in Drug Regulation[J]. China Food & Drug Administration Magazine(中国食品药品监管), 2025(3): 4-13.
[7] WU LH, GRAY M, DANG O, et al.RxBERT: Enhancing Drug Labeling Text Mining and Analysis with AI Language Modeling[J]. Experimental Biology and Medicine, 2024, 248(21): 1937-1943.
[8] CHOI JH, PALUMBO N, CHALASANI P, et al. MALADE: Orchestration of LLM-Powered Agents with Retrieval Augmented Generation for Pharmacovigilance[J/OL]. Cornell University arXiv, (2024-08-03)[2025-12-03]. https://arxiv.org/abs/2408.01869.
[9] NMPA. Notice from the General Department of the National Medical Products Administration on the List of Typical Application Scenarios of Artificial Intelligence in Drug Regulation[EB/OL]. (2024-06-18)[2025-12-03]. https://www.nmpa.gov.cn/xxgk/fgwj/gzwj/gzwjzh/20240618144318144.html.
[10] WU XR, LIU Y.Application Analysis of Large Language Models in Typical Drug Regulatory Scenarios[J]. China Food & Drug Administration Magazine(中国食品药品监管), 2025(9): 30-39.
[11] CHEN F, WU XR.Research on the Integrated Construction Strategy of Large Language Models in Drug Regulation[J]. Chinese Journal of Medicinal Guide(中国医药导刊), 2025, 27(1): 1-6.
[12] NMPA. Notice of the National Medical Products Administration on Revising Certain Provisions of the Drug Inspection Management Measures (Trial)[EB/OL]. (2023-07-21)[2025-10-30]. https://www.nmpa.gov.cn/xxgk/fgwj/gzwj/gzwjyp/20230721091201181.html.
[13] NMPA. Notice of the National Medical Products Administration on Guiding Principles for Risk Assessment of On-Site Inspection of Drug Production[EB/OL]. (2014-05-13)[2025-10-30]. https://www.nmpa.gov.cn/xxgk/fgwj/gzwj/gzwjyp/20140513120001350.html.
[14] WANG J, XIE T, LEI WF, et al.Overview of the Application of Large Language Models in Information Recognition[J]. Secrecy Science and Technology(保密科学技术), 2024(8): 52-57.
[15] HU SG, WU ZA.Research on GMP Inspection in China Based on Quality Risk Management[J]. Journal of Shenyang Pharmaceutical University(沈阳药科大学学报), 2024, 41(1): 121-127.
[16] HONG JM.Research on Application Skills of Excel Software Based on Functions[J]. Journal of Hunan Post and Telecommunication College(湖南邮电职业技术学院学报), 2025, 1(14): 22-27.
[17] HU XJ, WANG LJ, XU CB, et al.Analysis and Discussion of Typical Problems in On-Site Inspection of Biochemical Pharmaceutical Manufacturers in China[J]. China Pharmaceuticals(中国药业), 2024, 24(33): 34-38.
[18] General Office of the State Council. Opinions on Comprehensively Deepening the Reform of Drug and Medical Device Supervision to Promote High-Quality Development of the Pharmaceutical Industry[EB/OL]. (2025-01-30)[2025-10-30]. https://www.gov.cn/zhengce/content/202501/content_6996115.htm.
[19] WU YP, QIU XP, WANG YH, et al. Research on the Deconstruction of Causes and Governance of Information Pollution Triggered by the Hallucination of Large Language Models[J/OL]. Journal of Information (情报杂志), (2025-12-09)[2026-03-17]. https://link.cnki.net/urlid/61.1167.G3.20251208.0922.012.
[20] NMPA. Notice on the 14th Five-Year Plan for Drug Regulatory Network Security and Informatization Construction[EB/OL]. (2022-05-11)[2025-10-30]. https://www.nmpa.gov.cn/xxgk/fgwj/gzwj/gzwjzh/20220511110329171.html.