J. Robins, M. Hernán, and A. Rotnitzky. American journal of epidemiology, 166 (9):
994-1002; discussion 1003-4(November 2007)4950<m:linebreak></m:linebreak>GR: R01-HL080644/HL/NHLBI NIH HHS/United States; GR: R37-AI032475/AI/NIAID NIH HHS/United States; JID: 7910653; CON: Am J Epidemiol. 2007 Nov 1;166(9):985-93. PMID: 17875580; 2007/09/17 aheadofprint; ppublish;.
DOI: 10.1093/aje/kwm231
Abstract
Marginal structural models (MSMs) allow estimation of effect modification by baseline covariates, but they are less useful for estimating effect modification by evolving time-varying covariates. Rather, structural nested models (SNMs) were specifically designed to estimate effect modification by time-varying covariates. In their paper, Petersen et al. (Am J Epidemiol 2007;166:985-993) describe history-adjusted MSMs as a generalized form of MSM and argue that history-adjusted MSMs allow a researcher to easily estimate effect modification by time-varying covariates. However, history-adjusted MSMs can result in logically incompatible parameter estimates and hence in contradictory substantive conclusions. Here the authors propose a more restrictive definition of history-adjusted MSMs than the one provided by Petersen et al. and compare the advantages and disadvantages of using history-adjusted MSMs, as opposed to SNMs, to examine effect modification by time-dependent covariates.
%0 Journal Article
%1 Robins2007
%A Robins, James M
%A Hernán, Miguel A
%A Rotnitzky, Andrea
%D 2007
%J American journal of epidemiology
%K AntiretroviralTherapy CohortStudies ConfoundingFactors(Epidemiology) HIVInfections HIVInfections:drugtherapy HIVInfections:epidemiology HIVInfections:immunology HighlyActive HighlyActive:methods Humans LogisticModels LongitudinalStudies MathematicalComputing Models ProportionalHazardsModels Statistical TimeFactors TreatmentOutcome UnitedStates UnitedStates:epidemiology
%N 9
%P 994-1002; discussion 1003-4
%R 10.1093/aje/kwm231
%T Effect modification by time-varying covariates.
%U http://www.ncbi.nlm.nih.gov/pubmed/17875581
%V 166
%X Marginal structural models (MSMs) allow estimation of effect modification by baseline covariates, but they are less useful for estimating effect modification by evolving time-varying covariates. Rather, structural nested models (SNMs) were specifically designed to estimate effect modification by time-varying covariates. In their paper, Petersen et al. (Am J Epidemiol 2007;166:985-993) describe history-adjusted MSMs as a generalized form of MSM and argue that history-adjusted MSMs allow a researcher to easily estimate effect modification by time-varying covariates. However, history-adjusted MSMs can result in logically incompatible parameter estimates and hence in contradictory substantive conclusions. Here the authors propose a more restrictive definition of history-adjusted MSMs than the one provided by Petersen et al. and compare the advantages and disadvantages of using history-adjusted MSMs, as opposed to SNMs, to examine effect modification by time-dependent covariates.
%@ 0002-9262
@article{Robins2007,
abstract = {Marginal structural models (MSMs) allow estimation of effect modification by baseline covariates, but they are less useful for estimating effect modification by evolving time-varying covariates. Rather, structural nested models (SNMs) were specifically designed to estimate effect modification by time-varying covariates. In their paper, Petersen et al. (Am J Epidemiol 2007;166:985-993) describe history-adjusted MSMs as a generalized form of MSM and argue that history-adjusted MSMs allow a researcher to easily estimate effect modification by time-varying covariates. However, history-adjusted MSMs can result in logically incompatible parameter estimates and hence in contradictory substantive conclusions. Here the authors propose a more restrictive definition of history-adjusted MSMs than the one provided by Petersen et al. and compare the advantages and disadvantages of using history-adjusted MSMs, as opposed to SNMs, to examine effect modification by time-dependent covariates.},
added-at = {2023-02-03T11:44:35.000+0100},
author = {Robins, James M and Hernán, Miguel A and Rotnitzky, Andrea},
biburl = {https://www.bibsonomy.org/bibtex/2bd112214c676ff15a523f0a9f3363c13/jepcastel},
city = {Department of Epidemiology, Harvard School of Public Health, Boston, MA 02115, USA.},
doi = {10.1093/aje/kwm231},
interhash = {bee2ccabd85e9f90d27974a786cdec4c},
intrahash = {bd112214c676ff15a523f0a9f3363c13},
isbn = {0002-9262},
issn = {0002-9262},
journal = {American journal of epidemiology},
keywords = {AntiretroviralTherapy CohortStudies ConfoundingFactors(Epidemiology) HIVInfections HIVInfections:drugtherapy HIVInfections:epidemiology HIVInfections:immunology HighlyActive HighlyActive:methods Humans LogisticModels LongitudinalStudies MathematicalComputing Models ProportionalHazardsModels Statistical TimeFactors TreatmentOutcome UnitedStates UnitedStates:epidemiology},
month = {11},
note = {4950<m:linebreak></m:linebreak>GR: R01-HL080644/HL/NHLBI NIH HHS/United States; GR: R37-AI032475/AI/NIAID NIH HHS/United States; JID: 7910653; CON: Am J Epidemiol. 2007 Nov 1;166(9):985-93. PMID: 17875580; 2007/09/17 [aheadofprint]; ppublish;},
number = 9,
pages = {994-1002; discussion 1003-4},
pmid = {17875581},
timestamp = {2023-02-03T11:44:35.000+0100},
title = {Effect modification by time-varying covariates.},
url = {http://www.ncbi.nlm.nih.gov/pubmed/17875581},
volume = 166,
year = 2007
}