Objective  To construct an early warning and hierarchical response model for drug shortages.

Methods  To collect the drug use monitoring data and the shortage drug disposal data from public medical institutions in Beijing from January 2022 to December 2023. Based on the risk characteristics of regional and universal drug shortages, design shortage risk indicators and establish a regional risk early warning model using risk matrix evaluation. Establish a universal risk model using the Bayesian confidence propagation neural network (BCPNN) method, and further construct a drug shortage risk monitoring and hierarchical response model to verify the model's effectiveness through case analysis using relevant data of Beijing.

Results  In 2023, Beijing collected monitoring data on drug use from 402,618 cases, involving 10,872 varieties. The model identified drugs with universal shortage risks, such as methotrexate, nikethamide, and pralidoxime chloride, as well as drug varieties with regional shortage risks, including azithromycin (from Enterprise S), trimetazidine (from Enterprise J), and Maixuekang (from Enterprise C).The verification results showed that the regional model achieved a monitoring rate of 94.7% for the shortage drugs listed in the report of Beijing in 2023, and a monitoring rate of 72.7% for the shortage drugs not included in the list. The universal model had a monitoring rate of 31.6% for the shortage drugs reported in Beijing in 2023.

Conclusion  The monitoring results of the drug shortage risk early warning and hierarchical response model are quite consistent with the actual shortage situation, and the model has a certain degree of reliability and feasibility. It can provide a reference basis for addressing drug shortages.

1.杨露. 云南省短缺药品监测指标设计及处置体系构建研究[D]. 昆明: 昆明医科大学, 2022. DOI: 10.27202/d.cnki.gkmyc.2022.000006.

2.曲帅, 魏新江. 基于Logistic回归的短缺药品预警模型构建 [J]. 鲁东大学学报（自然科学版）, 2022, 38(4): 337-341. [Qu S, Wei XJ. Construction of drug shortage warning model based on logistic regression[J]. Ludong University Journal (Natural Science Edition), 2022, 38(4): 337-341.] DOI: 10.3969/j.issn.1673-8020.2022.04.007.

3.黄哲, 于钟怡, 兰毅鹏, 等. 辽宁省短缺药品监测评价模型的构建研究[J]. 沈阳药科大学学报, 2022, 39(8): 1010-1017. [Huang Z, Yu ZY, Lan YP, et al. Research on the construction of monitoring and evaluation model for shortage of drugs in Liaoning Province[J]. Journal of Shenyang Pharmaceutical University, 2022, 39(8): 1010-1017.] DOI: 10.14066/j.cnki.cn21-1349/r.2020.1090.

4.黄全元, 韦坤璇, 卢晓, 等. PDCA循环法在管理医院短缺药品中的应用[J]. 现代医药卫生, 2021, 37(3): 524-525, 536. [Huang QY, Wei KX, Lu X, et al. Application of the PDCA cycle method in managing shortages of medical supplies in hospitals[J]. Journal of Modern Medicine & Health, 2021, 37(3): 524-525, 536.] DOI: 10.3969/j.issn.1009-5519.2021.03.053.

5.马棪斐, 项程, 叶丹, 等. 基于网络监测数据的药品短缺研究[J]. 西北药学杂志, 2022, 37(6): 170-174. [Ma LF, Xiang C, Ye D, et al. Study of drug shortages based on the online monitoring data[J]. Northwest Pharmaceutical Journal, 2022, 37(6): 170-174.] DOI: 10.3969/j.issn.1004-2407.2022.06.032.

6.戴佳芯. 基于风险矩阵法的泸州市药品安全风险评估研究[D]. 重庆: 西南财经大学, 2020. DOI: 10.27412/d.cnki.gxncu.2020.002976.

7.肖彩云, 王奇巍, 刘悦, 等. 药品短缺预警量化指标研究综述[J]. 中国医药工业杂志, 2024, 55(4): 584-590. [Xiao CY, Wang  QW, Liu Y, et al. Review of quantitative indicators for early warning of drug shortage[J]. Chinese Journal of Pharmaceuticals, 2024, 55(4): 584-590.] DOI: 10.16522/j.cnki.cjph.2024.04.021.

8.曲帅. 集中采购背景下山东省药品短缺预警研究[D]. 山东烟台: 鲁东大学, 2023. DOI: 10.27216/d.cnki.gysfc.2023.000475.

9.傅政. Bayes方法筛选药物不良反应信号及利益风险研究[D]. 上海: 第二军医大学, 2008. https://cdmd.cnki.com.cn/Article/CDMD-90024-2008154253.htm.

10.李婵娟. 药品不良反应信号检测方法理论及应用研究[D]. 西安: 第四军医大学, 2008. DOI: 10.7666/d.d218023.

11.刘靖, 叶国菊, 王启明, 等. 模糊贝叶斯置信度递进神经网络法检测药品不良反应报告信号[J]. 中国药物警戒, 2022, 19(10): 1113-1117. [Liu J, Ye GJ, Wang QM, et al. Detection of report signals of adverse drug reactions by fuzzy BCPNN[J]. Chinese Journal of Pharmacovigilance, 2022, 19(10): 1113-1117.] DOI: 10.19803/j.1672-8629.20210189.

12.王蒙. 贝叶斯工具变量方法研究及其在药品不良反应主动监测中的应用[D]. 上海: 中国人民解放军海军军医大学, 2019. https://cdmd.cnki.com.cn/Article/CDMD-91020-1019123128.htm.

13.王蒙, 郭晓晶, 叶小飞, 等. 贝叶斯工具变量方法在药品安全性主动监测中的应用: 一项数据模拟研究[J]. 中国药物警戒, 2019, 16(3): 139-143. [Wang M, Guo XJ, Ye XF, et al. Application of bayesian instrumental variable in active surveillance data of adverse drug reactions:a simulation study[J]. Chinese Journal of Pharmacovigilance, 2019, 16(3): 139-143.] DOI: 10.3969/j.issn.1672-8629.2019.03.003.

14.洪东升, 倪剑, 单文雅, 等. 基于监测数据的药物不良反应快速识别及R语言实现[J]. 浙江大学学报（医学版）, 2020, 49(2): 253-259. [Hong DS, Ni J, Shan WY, et al. Establishment of a rapid identification of adverse drug reaction program in R language implementation based on monitoring data[J]. Journal of Zhejiang University (Medical Sciences) , 2020, 49(2): 253-259.] DOI: 10.3785/j.issn.1008-9292.2020.03.07.

15.王蓓丽, 赵丽, 张雪琴, 等. 基于真实世界数据的氟尿嘧啶注射剂不良反应信号挖掘与分析[J]. 药物流行病学杂志, 2025, 34(1): 12-20. [Wang BL, Zhao L, Zhang XQ, et al. Mining and analysis of adverse reaction signal of fluorouracil injection based on real-world data[J]. Chinese Journal of Pharmacoepidemiology, 2025, 34(1): 12-20.] DOI: 10.12173/j.issn.1005-0698.202408046.

16.何洁. 基于数据挖掘对疏血通注射液不良反应特点分析应用 [J]. 海峡药学, 2025, 37(2): 84-89. [He J. Analysis and application of adverse reaction characteristics of Shuxue-Tong injection based on data mining[J]. Strait Pharmaceutical Journal, 2025, 37(2): 84-89.] DOI: 10.3969/j.issn.1006-3765. 2025.02.023.

17.刘奕伶, 刘静琳, 卢珍珍, 等. 应用贝叶斯方法的抗感染领域研究现况分析[J]. 药物流行病学杂志, 2025, 34(1): 69-77. [Liu  YL, Liu JL, Lu ZZ, et al. Current status analysis of anti-infection research by using the Bayesian method[J]. Chinese Journal of Pharmacoepidemiology, 2025, 34(1): 69-77.] DOI: 10.12173/j.issn.1005-0698.202408053.

18.王紫灏. 基于FAERS数据库的质子泵抑制剂类药物不良反应数据挖掘研究[D]. 河南开封: 河南大学, 2024. DOI: 10.27114/d.cnki.ghnau.2024.002708.

19.周瑞珊, 卢佩雯, 陈君恒, 等. 药品不良反应数据挖掘技术在药物警戒中的应用[J]. 中国现代应用药学, 2024, 41(6): 864-870. [Zhou RS, Lu PW, Chen JH, et al. Application of adverse drug reaction data mining technology in drug vigilance[J]. Modern Applied Pharmacy in China, 2024, 41(6): 864-870.] DOI: 10.13748/j.cnki.issn1007-7693.20224098.

20.刘敏, 王海学. 药物临床试验不良反应因果关系评价方法概述与存在问题的思考[J]. 中国临床药理学杂志, 2023, 39(8): 1206-1211. [Liu M, Wang HX. Overview of causality evaluation methods for adverse reactions in drug clinical trials and consideration of existing problems[J]. The Chinese Journal of Clinical Pharmacology, 2023, 39(8): 1206-1211.] DOI: 10.13699/j.cnki.1001-6821.2023.08.030.

21.李婷, 张淼淼, 张田, 等. 伊布替尼单药或联合用药不良反应信号挖掘及影响因素分析[J]. 临床药物治疗杂志, 2023, 21(1): 64-70. [Li T, Zhang MM, Zhang T, et al. Analysis of adverse reaction signal mining and influencing factors of ibrutinib monotherapy or combination therapy[J]. Journal of Clinical Drug Therapy, 2023, 21(1): 64-70.] DOI: 10.3969/j.issn.1672-3384.2023.01.013.