Strategies and applications to address variable abundance differences in multicenter drug safety studies

Published on May. 30, 2023Total Views: 991 times Total Downloads: 379 times Download Mobile

Author: Yun-Xiao WU ¹ Lu XU ² Lin ZHUO ² Sheng-Feng WANG ¹  Si-Yan ZHAN ^1, ²

Affiliation: 1. Department of Epidemiology and Biostatistics, School of Public Health, Peking University, Beijing 100191, Chi-na 2. Research Center of Clinical Epidemiology, Peking University Third Hospital, Beijing 100191, China

Keywords: Drug safety Multi-center Variables missing Transfer learning

DOI: 10.19960/j.issn.1005-0698.202305012

Reference: Yun-Xiao WU, Lu XU, Lin ZHUO, Sheng-Feng WANG, Si-Yan ZHAN.Strategies and applications to address variable abundance differences in multicenter drug safety studies[J].Yaowu Liuxingbingxue Zazhi,2023, 32(5): 575-581.DOI: 10.19960/j.issn.1005-0698.202305012.[Article in Chinese]  Copied

Abstract
Full-text
References

Abstract

When adverse events of interest for post-marketing drug safety studies are relatively rare, conducting multi-center studies is necessary to address this issue. However, multi-center studies are often trapped in the problem of different central variables with varying degrees of richness, especially the complete lack of key variables in sub-centers, which causes studies not to make full use of all the information of each center when adjusting bias. Pro-pensity score calibration (PSC) and transfer learning proposed in recent years are available to deal with the complete absence of some variables in the sub-center. PSC has been applied to pharmacoepidemiology, but no report of transfer learning in this field has been published. This article will outline the characteristics and applications of the two methods, and focus on combing several types of transfer learning methods that can be used to solve such problems, and provide a reference for the in-depth study of transfer learning in multi-center drug safety evaluation.

Full-text

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

References

1.Li L, Yin J. Drug safety evaluation in China[J]. Curr Allergy Asthma Rep, 2019, 19(9): 39. DOI: 10.1007/s11882-019-0872-4.

2.StÜbner S, Grohmann R, Greil W, et al. Suicidal ideation and suicidal behavior as rare adverse events of antidepressant medication: current report from the AMSP multicenter drug safety surveillance pro-ject[J]. Int J Neuropsychopharmacol, 2018, 21(9): 814-821. DOI: 10.1093/ijnp/pyy048.

3.Van Domelen DR, Lyles RH. A look at the unique identifiability of propensity score calibration[J]. Am J Epidemiol, 2019, 188(7): 1397-1399. DOI: 10.1093/aje/kwz072.

4.Lunt M, Glynn RJ, Rothman KJ, et al. Propensity score calibration in the absence of surrogacy[J]. Am J Epidemiol, 2012, 175(12): 1294-1302. DOI: 10.1093/aje/kwr463.

5.Stürmer T, Schneeweiss S, Avorn J, et al. Adjusting effect estimates for unmeasured confounding with validation data using propensity score calibration[J]. Am J Epidemiol, 2005, 162(3): 279-289. DOI: 10.1093/aje/kwi192.

6.Han P. A further study of propensity score calibration in missing data analysis[J]. Statistica Sinica, 2018, 28(3): 1307-1332. DOI: 10.5705/ss.202016.0185.

7.Stürmer T, Glynn RJ, Rothman KJ, et al. Adjustments for unmeasured confounders in pharmacoepide-miologic database studies using external information[J]. Med Care, 2007, 45(10 Supl 2): S158-165. DOI: 10.1097/MLR.0b013e318070c045.

8.Wood ME, Frazier JA, Nordeng HM, et al. Prenatal triptan exposure and parent-reported early child-hood neurodevelopmental outcomes: an application of propensity score calibration to adjust for un-measured confounding by migraine severity[J]. Pharmacoepidemiol Drug Saf, 2016, 25(5): 493-502. DOI: 10.1002/pds.3902.

9.Thygesen LC, Pottegård A, Ersbøll AK, et al. External adjustment of unmeasured confounders in a case-control study of benzodiazepine use and cancer risk[J]. Br J Clin Pharmacol, 2017, 83(11): 2517-2527. DOI: 10.1111/bcp.13342.

10.Schneeweiss S. Sensitivity analysis and external adjustment for unmeasured confounders in epidemio-logic database studies of therapeutics[J]. Pharmacoepidemiol Drug Saf, 2006, 15(5): 291-303. DOI: 10.1002/pds.1200.

11.Zhuang F, Qi Z, Duan K, et al. A comprehensive survey on transfer learning[J]. Proceedings of the IEEE, 2021, 109(1): 43-76. DOI: 10.1109/JPROC.2020.3004555.

12.Chen D, Yang S, Zhou F. Transfer learning based fault diagnosis with missing data due to multi-rate sampling[J]. Sensors (Basel), 2019, 19(8): 1826. DOI: 10.3390/s19081826.

13.Pan SJ, Yang Q. A survey on transfer learning[J]. IEEE Trans Knowl Data Eng, 2010, 22(10): 1345-1359. DOI: 10.1109/TKDE.2009.191.

14.Atasever S, Azginoglu N, Terzi DS, et al. A comprehensive survey of deep learning research on medical image analysis with focus on transfer learning[J]. Clin Imaging, 2023, 94: 18-41. DOI: 10.1016/j.clinimag.2022.11.003.

15.黎英. 迁移学习在医学图像分析中的应用研究综述 [J]. 计算机工程与应用, 2021, 57(20): 42-52. [Li Y. Review of application of transfer learning in medical image analysis[J]. Computer Engineering and Applications, 2021, 57(20): 42-52.] DOI: 10.3778/j.issn.1002-8331.2106-0103.

16.Mei S, Zhu H. AdaBoost based multi-instance transfer learning for predicting proteome-wide interac-tions between Salmonella and human proteins[J]. PLoS One, 2014, 9(10): e110488. DOI: 10.1371/journal.pone.0110488.

17.Wang Q, Wang H, Wang L, et al. Diagnosis of chronic obstructive pulmonary disease based on trans-fer learning[J]. IEEE Access, 2020, 8: 47370-47383. DOI: 10.1109/ACCESS.2020.2979218.

18.Xu H, Pang J, Zhang W, et al. Predicting recurrence for patients with ischemic cerebrovascular events based on process discovery and transfer learning[J]. IEEE J Biomed Health Inform, 2021, 25(7): 2445-2453. DOI: 10.1109/JBHI.2021.3065427.

19.Sugiyama M, Suzuki T, Nakajima S, et al. Direct importance estimation for covariate shift adaptation[J]. Ann Inst Stat Math, 2008, 60(4): 699-746. DOI: 10.1007/s10463-008-0197-x.

20.Niu S, Hu Y, Wang J, et al. Feature-based distant domain transfer learning[A] //2020 IEEE International Conference on Big Data (Big Data)[C]. IEEE, 2020: 5164-5171.

21.Wiens J, Guttag J, Horvitz E. A study in transfer learning: leveraging data from multiple hospitals to enhance hospital-specific predictions[J]. J Am Med Inform Assoc, 2014, 21(4): 699-706. DOI: 10.1136/amiajnl-2013-002162.

22.韩普, 顾亮, 叶东宇, 等. 基于多任务和迁移学习的中文医学文献实体识别研究 [J/OL]. 数据分析与知识发现: 1-14. [2023-01-19] [Han P, Gu L, Ye DY, et al. Research on Chinese medical literature entity recognition based on multi-task and transfer learning[J]. Data Analysis and Knowledge Discovery: 1-14.] http://kns.cnki.net/kcms/detail/10.1478.G2.20221223.1648.011.html.

23.Puttagunta M, Ravi S. Medical image analysis based on deep learning approach[J]. Multimed Tools Appl, 2021, 80(16): 24365-24398. DOI: 10.1007/s11042-021-10707-4.

24.高爽, 徐巧枝. 迁移学习方法在医学图像领域的应用综述[J]. 计算机工程与应用, 2021, 57(24): 39-50. [Gao S, Xu QZ. Review of application of transfer learning in medical image field[J]. Computer Engineering and Applications, 2021, 57(24): 39-50.] DOI: 10.3778/j.issn.1002-8331.2107-0300.

25.Alzubaidi L, Al-Amidie M, Al-Asadi A, et al. Novel transfer learning approach for medical imaging with limited labeled data[J]. Cancers (Basel), 2021, 13(7): 1590. DOI: 10.3390/cancers13071590.

26.Hasib KM, Sakib S, Al Mahmud J, et al. Covid-19 prediction based on infected cases and deaths of bangladesh using deep transfer learning[A]//2022 IEEE World AI IoT Congress (AIIoT)[C]. IEEE, 2022: 296-302.

27.Liu K, Zhang X, Chen W, et al. Development and validation of a personalized model with transfer learning for acute kidney injury risk estimation using electronic health records[J]. JAMA Netw Open, 2022, 5(7): e2219776. DOI: 10.1001/jamanetworkopen.2022.19776.

28.Papernot N, Abadi M, Erlingsson U, et al. Semi-supervised knowledge transfer for deep learning from private training data[A] // 5th International Conference on Learning Representations[C]. IEEE, 2016. DOI: 10.48550/arXiv.1610.05755.