G方法简介。
An introduction to g methods.
作者信息
Naimi Ashley I, Cole Stephen R, Kennedy Edward H
机构信息
Department of Epidemiology, University of Pittsburgh.
Department of Epidemiology, University of North Carolina at Chapel Hill and.
出版信息
Int J Epidemiol. 2017 Apr 1;46(2):756-762. doi: 10.1093/ije/dyw323.
Abstract
Robins' generalized methods (g methods) provide consistent estimates of contrasts (e.g. differences, ratios) of potential outcomes under a less restrictive set of identification conditions than do standard regression methods (e.g. linear, logistic, Cox regression). Uptake of g methods by epidemiologists has been hampered by limitations in understanding both conceptual and technical details. We present a simple worked example that illustrates basic concepts, while minimizing technical complications.
摘要
罗宾斯广义方法（g方法）在比标准回归方法（如线性回归、逻辑回归、Cox回归）限制更少的识别条件下，提供了潜在结果对比（如差异、比率）的一致性估计。流行病学家对g方法的采用受到了在理解概念和技术细节方面的限制。我们给出一个简单的实例，以说明基本概念，同时尽量减少技术复杂性。
相似文献
1
An introduction to g methods.G方法简介。
Int J Epidemiol. 2017 Apr 1;46(2):756-762. doi: 10.1093/ije/dyw323.
2
Invited commentary: G-computation--lost in translation?特邀评论：G 计算——翻译中的迷失？
Am J Epidemiol. 2011 Apr 1;173(7):739-42. doi: 10.1093/aje/kwq474. Epub 2011 Mar 16.
3
Causal Methods for Observational Research: A Primer.观察性研究的因果方法：入门指南。
Arch Iran Med. 2018 Apr 1;21(4):164-169.
4
Differences between marginal structural models and conventional models in their exposure effect estimates: a systematic review.边缘结构模型与传统模型在暴露效应估计中的差异：系统评价。
Epidemiology. 2011 Jul;22(4):586-8. doi: 10.1097/EDE.0b013e31821d0507.
5
Response to invited commentary. Rose et al. respond to "G-computation and standardization in epidemiology".对特邀评论的回应。罗斯等人回应《流行病学中的G计算与标准化》。
Am J Epidemiol. 2011 Apr 1;173(7):743-4. doi: 10.1093/aje/kwq475.
6
Exploring the Subtleties of Inverse Probability Weighting and Marginal Structural Models.探索逆概率加权和边缘结构模型的细微差别。
Epidemiology. 2018 May;29(3):352-355. doi: 10.1097/EDE.0000000000000813.
7
Augmented and doubly robust G-estimation of causal effects under a Structural nested failure time model.结构嵌套失效时间模型下因果效应的增强型和双重稳健G估计
Biometrics. 2018 Jun;74(2):472-480. doi: 10.1111/biom.12749. Epub 2017 Jul 25.
8
G-computation demonstration in causal mediation analysis.因果中介分析中的G计算演示。
Eur J Epidemiol. 2015 Oct;30(10):1119-27. doi: 10.1007/s10654-015-0100-z. Epub 2015 Nov 4.
9
Marginal Structural Models: unbiased estimation for longitudinal studies.边缘结构模型：纵向研究的无偏估计。
Int J Public Health. 2011 Feb;56(1):117-9. doi: 10.1007/s00038-010-0198-4. Epub 2010 Oct 8.
10
Implementation of G-computation on a simulated data set: demonstration of a causal inference technique.在模拟数据集上实施 G 计算：因果推理技术的演示。
Am J Epidemiol. 2011 Apr 1;173(7):731-8. doi: 10.1093/aje/kwq472. Epub 2011 Mar 16.
引用本文的文献
1
Outcomes associated with use of makyokansekito, a Japanese herbal kampo medicine, in outpatients with community-acquired pneumonia: A retrospective cohort study.日本汉方药物麻子仁丸用于社区获得性肺炎门诊患者的相关结局：一项回顾性队列研究。
J Gen Fam Med. 2025 Jul 27;26(5):427-434. doi: 10.1002/jgf2.70052. eCollection 2025 Sep.
2
Treatment Comparison for a Single Arm Study Utilizing External Control: Performing Inference when Imputing Potential Outcomes.利用外部对照的单臂研究的治疗比较：在推算潜在结果时进行推断
Ther Innov Regul Sci. 2025 Jul 22. doi: 10.1007/s43441-025-00839-2.
3
Association between serial changes in serum C-reactive protein levels and mortality among patients with pleural empyema: a database research.胸腔积液患者血清C反应蛋白水平的系列变化与死亡率之间的关联：一项数据库研究
BMC Pulm Med. 2025 Jul 12;25(1):337. doi: 10.1186/s12890-025-03799-3.
4
Computing True Parameter Values in Simulation Studies Using Monte Carlo Integration.在模拟研究中使用蒙特卡罗积分计算真实参数值。
Epidemiology. 2025 Sep 1;36(5):690-693. doi: 10.1097/EDE.0000000000001873. Epub 2025 Jun 13.
5
Chronic kidney disease and dementia: an epidemiological perspective.慢性肾脏病与痴呆：流行病学视角
Nat Rev Nephrol. 2025 May 22. doi: 10.1038/s41581-025-00967-w.
6
M-Learning for Individual Treatment Rule With Survival Outcomes.用于具有生存结局的个体治疗规则的移动学习
Stat Med. 2025 May;44(10-12):e70093. doi: 10.1002/sim.70093.
7
Modelling diabetes and depression in Pakistan: using economic modelling to inform intervention design and a clinical trial of a behavioural activation intervention.巴基斯坦糖尿病与抑郁症的模型构建：运用经济模型为干预设计提供信息及一项行为激活干预的临床试验
BMJ Open. 2025 May 19;15(5):e092158. doi: 10.1136/bmjopen-2024-092158.
8
Flexible quantitative bias analysis for unmeasured confounding in subject-level indirect treatment comparisons with proportional hazards violation.针对比例风险违背情况下受试者水平间接治疗比较中未测量混杂因素的灵活定量偏倚分析。
BMC Med Res Methodol. 2025 May 10;25(1):131. doi: 10.1186/s12874-025-02551-z.
9
Biomarker-guided adaptive enrichment design with threshold detection for clinical trials with time-to-event outcome.用于具有事件发生时间结局的临床试验的、具有阈值检测的生物标志物引导的适应性富集设计。
J Biopharm Stat. 2025 Apr 20:1-18. doi: 10.1080/10543406.2025.2489291.
10
Estimating treatment effects using parametric models as counter-factual evidence.使用参数模型估计治疗效果作为反事实证据。
BMC Med Res Methodol. 2025 Apr 9;25(1):91. doi: 10.1186/s12874-025-02540-2.
本文引用的文献
1
Causal Inference Under Multiple Versions of Treatment.多种治疗版本下的因果推断
J Causal Inference. 2013 May 1;1(1):1-20. doi: 10.1515/jci-2012-0002.
2
Occupational radon exposure and lung cancer mortality: estimating intervention effects using the parametric g-formula.职业性氡暴露与肺癌死亡率：使用参数化g公式估计干预效果
Epidemiology. 2014 Nov;25(6):829-34. doi: 10.1097/EDE.0000000000000164.
3
The parametric g-formula for time-to-event data: intuition and a worked example.用于事件发生时间数据的参数化g公式：直观理解与实例分析。
Epidemiology. 2014 Nov;25(6):889-97. doi: 10.1097/EDE.0000000000000160.
4
Analysis of occupational asbestos exposure and lung cancer mortality using the g formula.应用 g 公式分析职业性石棉暴露与肺癌死亡率。
Am J Epidemiol. 2013 May 1;177(9):989-96. doi: 10.1093/aje/kws343. Epub 2013 Apr 4.
5
Methods for dealing with time-dependent confounding.处理时依混杂的方法。
Stat Med. 2013 Apr 30;32(9):1584-618. doi: 10.1002/sim.5686. Epub 2012 Dec 3.
6
The parametric g-formula to estimate the effect of highly active antiretroviral therapy on incident AIDS or death.评估高效抗逆转录病毒疗法对艾滋病事件或死亡影响的参数 g 公式。
Stat Med. 2012 Aug 15;31(18):2000-9. doi: 10.1002/sim.5316. Epub 2012 Apr 11.
7
Differences between marginal structural models and conventional models in their exposure effect estimates: a systematic review.边缘结构模型与传统模型在暴露效应估计中的差异：系统评价。
Epidemiology. 2011 Jul;22(4):586-8. doi: 10.1097/EDE.0b013e31821d0507.
8
Invited commentary: positivity in practice.特邀评论：实践中的积极性。
Am J Epidemiol. 2010 Mar 15;171(6):674-7; discussion 678-81. doi: 10.1093/aje/kwp436. Epub 2010 Feb 5.
9
Illustrating bias due to conditioning on a collider.图示由于在共因上进行条件推断而产生的偏差。
Int J Epidemiol. 2010 Apr;39(2):417-20. doi: 10.1093/ije/dyp334. Epub 2009 Nov 19.
10
Intervening on risk factors for coronary heart disease: an application of the parametric g-formula.干预冠心病的危险因素：参数 g 公式的应用。
Int J Epidemiol. 2009 Dec;38(6):1599-611. doi: 10.1093/ije/dyp192. Epub 2009 Apr 23.
Zotero 插件