Guide on Methodological Standards in Pharmacoepidemiology (2nd edition) and their series interpretation (17): application of predictive models in drug effectiveness and safety evaluation

Published on May. 28, 2026Total Views: 80 times Total Downloads: 23 times Download Mobile

Author: NIE Xiaolu ¹ WU Yunxiao ^2, ³ ZHAO Houyu ⁴  ZHAN Siyan ^2, ^3, ^4, ^5, ⁶  SUN Feng ^2, ^3, ^5, ^6, ^7, ^8, ⁹

Affiliation: 1. Center for Clinical Epidemiology & Evidence-based Medicine, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing 100045, China 2. Department of Epidemiology and Biostatistics, School of Public Health, Peking University, Beijing 100191, China 3. Key Laboratory of Epidemiology of Major Diseases (Peking University), Ministry of Education, Beijing 100191, China 4. Research Center of Clinical Epidemiology, Peking University Third Hospital, Beijing 100191, China 5. Center of Postmarketing Safety Evaluation, Peking University Health Science Center, Beijing 100191, China 6. Center for Intelligent Public Health, Institute for Artificial Intellignce, Peking University, Beijing 100191, China 7. Department of Ophthalmology, Peking University Third Hospital, Beijing 100191, China 8. School of Traditional Chinese Medicine, Xinjiang Medical University, Urumqi 830017, China 9. School of Public Health, Shihezi University, Shihezi 832000, Xinjiang Uygur Autonomous Region, China

Keywords: Predictive models Pharmacoepidemiology Precision Medicine Ef-fectiveness evaluation Safety evaluation Heterogeneity of treatment effects Biomarkers

DOI: 10.12173/j.issn.1005-0698.202605004

Reference: Nie Xl, Wu YX, Zhao HY, et al. Guide on Methodological Standards in Pharmacoepidemiology (2nd edition)and their series interpretation (17): application of predictive models in drug effectiveness and safety evaluation[J]. Chinese Journal of Pharmacoepidemiology, 2026, 35(5): 481-492. DOI: 10.12173/j.issn.1005-0698.202605004. [Article in Chinese]  Copied

Abstract
Full-text
References

Abstract

Predictive models are essential methodological tools for individualized therapy and precision medicine. Based on the Guide on Methodological Standards in Pharmacoepidemiology (2nd Edition), this paper systematically elucidates four core application scenarios of predictive models in drug effectiveness and safety evaluation: heterogeneity of treatment effects identification (who should receive the drug), biomarker discovery (why it is effective), effectiveness prediction for patients receiving drugs (how effective it is), and drug safety evaluation (whether it is safe). Furthermore, four special methodological considerations are discussed: the essential distinction between predictive and causal models, strategies for handling class imbalance and rare event prediction, the three-level translational pathway from prediction to decision-making (internal validation, external validation, and clinical utility assessment), and the closed-loop system for reporting specification and bias risk assessment formed by TRIPOD+AI and PROBAST+AI. This paper aims to provide methodological guidance for pharmacoepidemiologists and facilitate paradigm shift from "model development" to "clinical utility" in this field.

Full-text

Please download the PDF version to read the full text: download

References

1.颜济南, 吴昀效, 聂晓璐, 等. 《中国药物流行病学研究方法学指南(第2版)》的制订/修订过程[J]. 药物流行病学杂志, 2025,34(2):121-135. [Yan JN, Wu YX, Nie XL, et al. Revision process of the Guide on Methodological Standards in Pharmacoepidemiology in China (2nd edition)[J]. Chinese Journal of Pharmacoepidemiology, 2025, 34(2):121-135.] DOI: 10.12173/j.issn.1005-0698.202502028.

2.Mant D. Can randomised trials inform clinical decisions about individual patients? [J]. Lancet, 1999, 353(9154): 743-746. DOI: 10.1016/S0140-6736(98)09102-8.

3.Kent DM, Hayward RA. Limitations of applying summary results of clinical trials to individual patients: the need for risk stratification[J]. JAMA, 2007, 298(10):1209-1212. DOI: 10.1001/jama.298.10.1209.

4.Rothwell PM. Treating individuals 2. Subgroup analysis in randomised controlled trials: importance, indications, and inter-pretation[J]. Lancet, 2005, 365(9454):176-186. DOI: 10.1016/S0140-6736(05)17709-5.

5.Kent DM, Steyerberg E, van Klaveren D. Personalized evidence based medicine: predictive approaches to heterogeneous treat-ment effects[J]. BMJ, 2018, 363: k4245. DOI: 10.1136/bmj.k4245.

6.Kent DM, Paulus JK, van Klaveren D, et al. The Predictive Appr-oaches to Treatment effect Heterogeneity (PATH) Statement[J]. Ann Intern Med, 2020, 172(1): 35-45. DOI: 10.7326/M18-3667.

7.Kent DM, van Klaveren D, Paulus JK, et al. The Predictive Approaches to Treatment effect Heterogeneity (PATH) Statement: Explanation and Elaboration[J]. Ann Intern Med, 2020, 172(1):W1-W25. DOI: 10.7326/m18-3668.

8.Kumar V, Shaw JR, Key NS, et al. D-dimer enhances risk-targeted thromboprophylaxis in ambulatory patients with cancer[J]. Oncologist, 2020, 25(12):1075-1083. DOI: 10.1002/onco. 13540.

9.Rekkas A, van Klaveren D, Ryan PB, et al. A standardized framework for risk-based assessment of treatment effect heter-ogeneity in observational healthcare databases[J]. NPJ Digit Med, 2023, 6(1): 58. DOI: 10.1038/s41746-023-00794-y.

10.Xu Y, Huang T, Zhang Y, et al. Risk of heart failure hospitalization for GLP-1 receptor agonists versus DPP-4 inhi-bitors or SGLT-2 inhibitors in patients with type 2 diabetes: a target trial emulation[J]. Circulation, 2026, 153(15):1086-1100. DOI: 10.1161/CIRCULATIONAHA.125.075157.

11.张伟霞. 基于药物基因组学的药物流行病学研究设计概述[J]. 药物流行病学杂志, 2021, 30(12):787-792. [Zhang WX. Overview of pharmacoepidemiological research design based on pharmacogenomics[J]. Chinese Journal of Pharmacoepidemiology, 2021, 30(12):787-792.] DOI: 10.19960/j.cnki.issn1005-0698.2021.12.001.

12.Uk Biobank. Molecular pharmacoepidemiology: the assessment of the effectiveness and safety of commonly prescribed medications in routine clinical practice according to genetic variability [EB/OL]. (2025-03-18) [2026-04-12]. https://www.ukbiobank.ac.uk.

13.Ran M, Zhang SL, Tam KY. Identifying meaningful drug respo-nse biomarkers from public pharmacogenomic datasets with biologically informed interpretable neural networks[J]. Comput Biol Chem, 2026, 120(Pt 1):108669. DOI: 10.1016/j.compbiolchem.2025.108669.

14.Li Y, Ding J, Du F, et al. Deep learning-based multimodal prediction of NAC response in LARC by integrating MRI and pr-oteomics[J]. Cancer Res Treat, 2025, DOI: 10.4143/crt.2025.707.

15.Wang Y, Liu R, Zhang Y, et al. A network-driven framework for drug response precision prediction of acute myeloid leukemia[J]. Adv Sci (Weinh), 2025, 12(36):e06447. DOI: 10.1002/advs. 202506447.

16.Morikubo H, Tojima R, Maeda T, et al. Machine learning using clinical data at baseline predicts the medium-term efficacy of ustekinumab in patients with ulcerative colitis[J]. Sci Rep, 2024, 14(1): 4386. DOI: 10.1038/s41598-024-55126-1.

17.Park J, Kim D, Lee D, et al. DeepTDM: deep learning-based prediction of sequential therapeutic drug monitoring levels of vancomycin[J]. IEEE J Transl Eng Health Med, 2025, 13:493-506. DOI: 10.1109/JTEHM.2025.3623605.

18.聂晓璐, 赵厚宇, 霍东辉, 等. 基于真实世界数据与定量构效关系的儿童药源性免疫性血小板减少症研究[J]. 中国药物警戒, 2025, 22(1):16-22. [Nie XL, Zhao HY, Huo DH, et al. Pediatric drug-induced immune thrombocytopenia based on real world data and quantitative structure-activity relationship[J]. Chinese Journal of Pharmacovigilance, 2025, 22(1):16-22.] DOI: 10.19803/j.1672-8629.20240807.

19.聂晓璐, 孙凤, 阎爱侠, 等. 基于多源信息构建药品不良反应基准数据库[J]. 中国药物警戒, 2025, 22(1): 10-15, 36. [Nie XL, Sun F, Yan AX, et al. Methodology for constructing benchmark database of adverse drug reactions based on multi-source information[J]. Chinese Journal of Pharmacovigilance, 2025, 22(1): 10-15, 36.] DOI: 10.19803/j.1672-8629.20240911.

20.姚克宇, 张舒琪, 金锐, 等. 药物致肝损伤数据集的比较研究[J]. 中国药物警戒, 2023, 20(5):568-573. [Yao KY, Zhang SQ, Jin R, et al. Comparative study on data set of drug - induced liver injury[J]. Chinese Journal of Pharmacovigilance, 2023, 20(5): 568-573.] DOI: 10.19803/j.1672-8629.20220447.

21.Connor S, Li T, Qu Y, et al. Generation of a drug-induced renal injury list to facilitate the development of new approach methodologies for nephrotoxicity[J]. Drug Discov Today, 2024, 29(4): 103938. DOI: 10.1016/j.drudis.2024.103938.

22.Qu Y, Li T, Liu Z, et al. DICTrank is a reliable dataset for cardiotoxicity prediction using machine learning methods[J]. Chem Res Toxicol, 2025, 38(4):647-655. DOI: 10.1021/acs.chemrestox.4c00428.

23.US Food and Drug Administration. SafetAI Initiative[EB/OL]. (2025-09-05)[2026-04-13]. https://www.fda.gov/about-fda/nctr-research-focus-areas/safetai-initi-ative.

24.Li T, Tong W, Roberts R, et al. DeepCarc: deep learning-powered carcinogenicity prediction using model-level represe-ntation[J]. Front Artif Intell, 2021, 4:757780. DOI: 10.3389/frai.2021.757780.

25.Li T, Liu Z, Thakkar S, et al. DeepAmes: a deep learning-powered Ames test predictive model with potential for regulatory application[J]. Regul Toxicol Pharmacol, 2023, 144: 105486. DOI: 10.1016/j.yrtph.2023.105486.

26.刘溪, 李晨, 田源, 等. 基于自发报告数据驱动机器学习模型预测严重药品不良反应[J]. 中国药物警戒, 2025, 22(12): 1410-1417. [Liu X, Li C, Tian Y, et al. Prediction of severe adverse drug reactions based on spontaneous reporting data-driven machine learning models[J]. Chinese Journal of Pharmaco-vigilance, 2025, 22(12):1410-1417.] DOI: 10.19803/j.1672-8629.20250512.

27.NMPA. Measures for the administration of reporting and monitoring of adverse drug reactions [EB/OL]. (2011-05-04) [2026-04-25]. https://www.nmpa.gov.cn/yaopin/ypfgwj/ypfgbmgzh/20110504162501325.html.

28.National Cancer Institute. Common Terminology Criteria for Adverse Events (CTCAE) Version 5.0[EB/OL].(2017-11-27)[2026-04-26]. https://dctd.cancer.gov/research/ctep-trials/for-sites/adverse-events/ctcae-v5-5x7.pdf.

29.江永贤, 赵珊, 李张悦, 等. 人工智能技术在药物不良事件预测中的应用现状及模型性能评估:一项系统文献分析[J]. 药品不良反应杂志, 2025, 27(12):735-744. [Jiang YX, Zhao S, Li ZY, et al. Application status of artificial intelligence technology in adverse drug event prediction and model performance evaluation: a systematic literature analysis[J]. Adverse Drug Reactions Journal, 2025, 27(12):735-744.] DOI: 10.3760/cma.j.cn114015-20241211-00200.

30.Kim JV, Davis SE, Matheny ME, et al. Integrating electronic health records with other data sources for postmarket drug safety signal identification: a review[J]. Front Drug Saf Regul, 2024, 4:1428831. DOI: 10.3389/fdsfr.2024.1428831.

31.Li F, Rasmy L, Xiang Y, et al. Dynamic prognosis prediction for patients on DAPT after drug-eluting stent implantation: model development and validation[J]. J Am Heart Assoc, 2024, 13(3):e029900. DOI: 10.1161/JAHA.123.029900.

32.龙昱军, 刘剑钊, 杨智荣. 机器学习在多重用药不良事件预测中的应用:典型方法与案例的综述[J]. 药物流行病学杂志, 2026, 35(3):327-335. [Long YJ, Liu JZ, Yang ZR. Machine learning for predicting adverse drug events in polypharmacy: a review of methods and case studies[J]. Chinese Journal of Pharmacoepidemiology, 2026, 35(3): 327-335.] DOI: 10.12173/j.issn.1005-0698.202505005.

33.Abécassis J, Dumas É, Alberge J, et al. From prediction to prescription: machine learning and causal inference for the heterogeneous treatment Effect[J]. Annual Review Biomedical Data Science, 2025, 8: 381-404. DOI: 10.1146/annurev-biodatasci-103123-095750.

34.Feuerriegel SAO, Frauen D, Melnychuk V, et al. Causal machine learning for predicting treatment outcomes[J]. Nat Med, 2024, 30: 958-968. DOI: 10.1038/s41591-024-02902-1.

35.周江杰, 王予童, 冯天, 等. 大数据驱动的罕见事件非均衡数据分析方法研究进展[J]. 药物流行病学杂志, 2025, 34(8): 952-961. [Zhou JJ, Wang YT, Feng T, et al. Research progress on big-data-driven analysis strategies for imbalanced data of rare events[J]. Chinese Journal of Pharmacoepidemiology, 2025, 34(8): 952-961.] DOI: 10.12173/j.issn.1005-0698.202411080.

36.Chalabianloo N, Ahmadi F, Omrani MA, et al. Machine learning methods for predicting adverse drug events: a systematic review[J]. Br J Clin Pharmacol, 2026, 92(2):422-444. DOI: 10.1002/bcp.70377.

37.Collins GAO, Dhiman P, Ma J, et al. Evaluation of clinical prediction models (part 1): from development to external validation[J]. BMJ, 2024, 384:e074819. DOI: 10.1136/bmj-2023-074819.

38.Amusa T, Okunola D, Izinyon O, et al. Strategies for embedding prediction models in clinical decision making workflows[J]. Cureus, 2026, 18(1): e101185. DOI: 10.7759/cureus.101185.

39.Collins GAO, Moons KAOX, Dhiman PAO, et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods[J]. BMJ, 2024, 385: e078378. DOI: 10.1136/bmj-2023-078378.

40.Cohen JF, Bossuyt PMM. TRIPOD+AI: an updated reporting guideline for clinical prediction models[J]. BMJ, 2024, 385: q824. DOI: 10.1136/bmj.q824.

41.Xu J, Guo Y, Wang F, et al. Protocol for the development of a reporting guideline for causal and counterfactual prediction models in biomedicine[J]. BMJ Open, 2022, 12(6): e059715. DOI: 10.1136/bmjopen-2021-059715.

42.Moons KGM, Wolff RF, Riley RD, et al. PROBAST: a tool to assess risk of bias and applicability of prediction model studies: explanation and elaboration[J]. Ann Intern Med, 2019, 170(1):W1-W33. DOI: 10.7326/M18-1377.

43.Moons KAOX, Damen JAO, Kaul TAO, et al. PROBAST+AI: an updated quality, risk of bias, and applicability assessment tool for prediction models using regression or artificial intelligence methods[J]. BMJ, 2025, 388: e082505. DOI: 10.1136/bmj-2024-082505.