贝叶斯网络荟萃分析中的随机临床试验二元结果综述
- 作者: Sapozhnikov K.V.1, Parfenov S.A.1, Lazarev A.A.2, Kirichek R.V.2, Tolkacheva D.G.3, Mironenko O.N.3, Klishkova N.V.1, Kulishenko V.V.1
-
隶属关系:
- Kirov Military Medical Academy
- Saint Petersburg State University of Telecommunications
- Russian Presidential Academy of National Economy and Public Administration
- 期: 卷 26, 编号 3 (2024)
- 页面: 473-482
- 栏目: Discussion
- URL: https://journals.rcsi.science/1682-7392/article/view/264268
- DOI: https://doi.org/10.17816/brmma629333
- ID: 264268
如何引用文章
详细
本文介绍了贝叶斯网络元分析作为一种利用数学模型进行间接比较的方法的主要方面。为了说明贝叶斯网络荟萃分析法的运行情况,引入了随机效应和固定效应两种主要模型的程序代码。这些模型由 Pascal 组件编写,并在 JAGS 程序中运行。用于建模的数据是从 D. Hu、 A.M. O'Connor, S. Wang 等人的论文[6]中生成的数据,以便核对结果。使用 R 语言和 rjags 软件包将数据载入模型并运行程序。在确定最佳模型时,使用了原始 R 代码计算的模型充分性指标,如总残差、杠杆率和信息偏差标准。此外,还使用 ggplot2 软件包确定模型充分性的图形方法。在考虑到过渡性和异质性假设的情况下,以临床试验中现有的药物疗效结果为基础构建证据网络的例子进行了探讨。概述了进行间接比较和直接比较以确定药物真实估计值的可能性。还解释了贝叶斯统计的要素,如先验概率、后验概率和可能性,以及在荟萃分析中使用这些要素的优势。介绍了广义线性模型的一般和特殊形式的数学装置,用于使用二项式输出来获得治疗效果的相对估计。对模型的性能进行了解释。在对充分性指标进行比较后,随机效应模型比固定效应模型更具优势。为了达到更好的充分性,应该花时间仔细卸载出版物中的数据,并选择有参考价值的先验指标。总的来说,贝叶斯综合分析是一种独特而重要的网络荟萃分析。它的特殊之处在于使用概率方法进行数据分析。了解了解贝叶斯统计的基本原理也是在各个研究领域成功使用这种方法的一个重要方面。然而,要有效地应用这种方法,必须注意仔细的数据准备和先验表示的选择。与其他荟萃分析方法相比,如果有信息丰富的先验分布和适当的实施,贝叶斯综合法可以提供更准确、更可靠的结果。贝叶斯综合法是世界和俄罗斯联邦公认的统计数据分析方法。
作者简介
Kirill V. Sapozhnikov
Kirov Military Medical Academy
编辑信件的主要联系方式.
Email: vmeda-nio@mil.ru
ORCID iD: 0000-0002-2476-7666
SPIN 代码: 2707-0339
Scopus 作者 ID: 57200810332
Researcher ID: ААЕ-3453-2022
MD, Cand. Sci. (Med.)
俄罗斯联邦, Saint PetersburgSergei A. Parfenov
Kirov Military Medical Academy
Email: sa.parfenov1988@yandex.ru
ORCID iD: 0000-0002-1649-9796
SPIN 代码: 6939-6910
MD, Cand. Sci. (Med.)
俄罗斯联邦, Saint PetersburgAndrei A. Lazarev
Saint Petersburg State University of Telecommunications
Email: Andrey.05.03.ru@mail.ru
ORCID iD: 0009-0006-6204-8423
SPIN 代码: 9715-2124
graduate student
俄罗斯联邦, Saint PetersburgRuslan V. Kirichek
Saint Petersburg State University of Telecommunications
Email: kirichek@sut.ru
ORCID iD: 0000-0002-8781-6840
SPIN 代码: 3253-4972
Dr. Sci. (Tech.), associate professor
俄罗斯联邦, Saint PetersburgDaria G. Tolkacheva
Russian Presidential Academy of National Economy and Public Administration
Email: tolkacheva.d@gmail.com
ORCID iD: 0000-0002-6314-4218
SPIN 代码: 4186-5243
Scopus 作者 ID: 57221817074
independent expert of research projects
俄罗斯联邦, MoscowOlga N. Mironenko
Russian Presidential Academy of National Economy and Public Administration
Email: freelomir@yandex.ru
ORCID iD: 0000-0001-8952-8386
SPIN 代码: 3265-8708
Cand. Sci. (Econ.)
俄罗斯联邦, MoscowNatalia V. Klishkova
Kirov Military Medical Academy
Email: N-Klishkova@yandex.ru
ORCID iD: 0000-0003-0273-0931
SPIN 代码: 7031-7397
Cand. Sci. (Ped.), associate professor
俄罗斯联邦, Saint PetersburgValeriy V. Kulishenko
Kirov Military Medical Academy
Email: v_kulishenko@mail.ru
ORCID iD: 0000-0002-3872-3357
SPIN 代码: 1899-7341
MD, Cand. Sci. (Med.)
俄罗斯联邦, Saint Petersburg参考
- Bogdanov AA, Bogdanov AA. Network meta-analysis for clinicians. Modern Oncology. 2021;23(3):418–424. EDN: IVQHBW doi: 10.26442/18151434.2021.3.201202
- Omelianovskii VV, Avksenteva MV, Sura MV, et al. Methodological recommendations for conducting meta-analyses. Moscow: Center for Expertise and Quality Control of Medical Care; 2017. 28 р. (In Russ.)
- Sadeghirad B, Foroutan F, Zoratti MJ, et al. Theory and practice of Bayesian and frequentist frameworks for network meta-analysis. BMJ Evid Based Med. 2023;28(3):204–209. doi: 10.1136/bmjebm-2022-111928
- Li T, Puhan MA, Vedula SS, Singh S, et al. Network meta-analysis-highly attractive but more methodological research is needed. BMC Med. 2011;9:79. doi: 10.1186/1741-7015-9-79
- Chaimani A, Caldwell DM, Li T, et al. Undertaking network meta-analyses. Cochrane handbook for systematic reviews of interventions. 2019:285–320. doi: 10.1002/9781119536604.ch11
- Hu D, O’Connor AM, Wang C, et al. How to conduct a Bayesian network meta-analysis. Front Vet Sci. 2020;(7):271. doi: 10.3389/fvets.2020.00271
- Rouse B, Chaimani A, Li T. Network meta-analysis: an introduction for clinicians. Intern Emerg Med. 2017;12:103–111. doi: 10.1007/s11739-016-1583-7
- Bland JM, Altman DG. Bayesians and frequentists. BMJ. 1998;317(7166):1151–1160. doi: 10.1136/bmj.317.7166.1151
- Kruschke JK, Liddell TM. The Bayesian New Statistics: Hypothesis testing, estimation, meta-analysis, and power analysis from a Bayesian perspective. Psychon Bull Rev. 2018;25:178–206. doi: 10.3758/s13423-016-1221-4
- Hespanhol L, Vallio CS, Costa LM, Saragiotto BT. Understanding and interpreting confidence and credible intervals around effect estimates. Braz J Phys Ther. 2019;23(4):290–301. doi: 10.1016/j.bjpt.2018.12.006
- Dias S, Ades AE, Welton NJ, Jansen JP. Network meta-analysis for decision-making. Bristol: John Wiley & Sons; 2018. 473 р.
- Rosenberger KJ, Xing A, Murad MH, et al. Prior choices of between-study heterogeneity in contemporary Bayesian network meta-analyses: an empirical study. J Gen Intern Med. 2021;36(4):1049–1057. doi: 10.1007/s11606-020-06357-1
- White IR, Turner RM, Karahalios A, Salanti G. A comparison of arm-based and contrast-based models for network meta-analysis. Stat Med. 2019;38(27):5197–5213. doi: 10.1002/sim.8360
- Lin L, Chu H, Hodges JS. On evidence cycles in network meta-analysis. Stat Interface. 2020;13(4):425. doi: 10.4310/sii.2020.v13.n4.a1
- Dias S, Welton NJ, Sutton AJ, Ades AE. NICE DSU technical support document 2: a generalised linear modelling framework for pairwise and network meta-analysis of randomised controlled trials. Bristol: National Institute for Health and Clinical Excellence; 2011. 98 р.