Advanced search
1 file | 433.89 KB Add to list

Analysis of longitudinal studies with repeated outcome measures : adjusting for time-dependent confounding using conventional methods

(2018) AMERICAN JOURNAL OF EPIDEMIOLOGY. 187(5). p.1085-1092
Author
Organization
Abstract
Estimation of causal effects of time-varying exposures using longitudinal data is a common problem in epidemiology. When there are time-varying confounders, which may include past outcomes, affected by prior exposure, standard regression methods can lead to bias. Methods such as inverse probability weighted estimation of marginal structural models have been developed to address this problem. However, in this paper we show how standard regression methods can be used, even in the presence of time-dependent confounding, to estimate the total effect of an exposure on a subsequent outcome by controlling appropriately for prior exposures, outcomes, and time-varying covariates. We refer to the resulting estimation approach as sequential conditional mean models (SCMMs), which can be fitted using generalized estimating equations. We outline this approach and describe how including propensity score adjustment is advantageous. We compare the causal effects being estimated using SCMMs and marginal structural models, and we compare the two approaches using simulations. SCMMs enable more precise inferences, with greater robustness against model misspecification via propensity score adjustment, and easily accommodate continuous exposures and interactions. A new test for direct effects of past exposures on a subsequent outcome is described.
Keywords
direct effect, indirect effect, inverse probability weight, longitudinal study, marginal structural model, sequential conditional mean model, time-varying confounder, total effect, MARGINAL STRUCTURAL MODELS, INVERSE PROBABILITY WEIGHTS, CAUSAL INFERENCE, G-COMPUTATION, ADJUSTMENT

Downloads

  • kwx311.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 433.89 KB

Citation

Please use this url to cite or link to this publication:

MLA
Keogh, Ruth H et al. “Analysis of Longitudinal Studies with Repeated Outcome Measures : Adjusting for Time-dependent Confounding Using Conventional Methods.” AMERICAN JOURNAL OF EPIDEMIOLOGY 187.5 (2018): 1085–1092. Print.
APA
Keogh, R. H., Daniel, R. M., VanderWeele, T. J., & Vansteelandt, S. (2018). Analysis of longitudinal studies with repeated outcome measures : adjusting for time-dependent confounding using conventional methods. AMERICAN JOURNAL OF EPIDEMIOLOGY, 187(5), 1085–1092.
Chicago author-date
Keogh, Ruth H, Rhian M Daniel, Tyler J VanderWeele, and Stijn Vansteelandt. 2018. “Analysis of Longitudinal Studies with Repeated Outcome Measures : Adjusting for Time-dependent Confounding Using Conventional Methods.” American Journal of Epidemiology 187 (5): 1085–1092.
Chicago author-date (all authors)
Keogh, Ruth H, Rhian M Daniel, Tyler J VanderWeele, and Stijn Vansteelandt. 2018. “Analysis of Longitudinal Studies with Repeated Outcome Measures : Adjusting for Time-dependent Confounding Using Conventional Methods.” American Journal of Epidemiology 187 (5): 1085–1092.
Vancouver
1.
Keogh RH, Daniel RM, VanderWeele TJ, Vansteelandt S. Analysis of longitudinal studies with repeated outcome measures : adjusting for time-dependent confounding using conventional methods. AMERICAN JOURNAL OF EPIDEMIOLOGY. 2018;187(5):1085–92.
IEEE
[1]
R. H. Keogh, R. M. Daniel, T. J. VanderWeele, and S. Vansteelandt, “Analysis of longitudinal studies with repeated outcome measures : adjusting for time-dependent confounding using conventional methods,” AMERICAN JOURNAL OF EPIDEMIOLOGY, vol. 187, no. 5, pp. 1085–1092, 2018.
@article{8591112,
  abstract     = {Estimation of causal effects of time-varying exposures using longitudinal data is a common problem in epidemiology. When there are time-varying confounders, which may include past outcomes, affected by prior exposure, standard regression methods can lead to bias. Methods such as inverse probability weighted estimation of marginal structural models have been developed to address this problem. However, in this paper we show how standard regression methods can be used, even in the presence of time-dependent confounding, to estimate the total effect of an exposure on a subsequent outcome by controlling appropriately for prior exposures, outcomes, and time-varying covariates. We refer to the resulting estimation approach as sequential conditional mean models (SCMMs), which can be fitted using generalized estimating equations. We outline this approach and describe how including propensity score adjustment is advantageous. We compare the causal effects being estimated using SCMMs and marginal structural models, and we compare the two approaches using simulations. SCMMs enable more precise inferences, with greater robustness against model misspecification via propensity score adjustment, and easily accommodate continuous exposures and interactions. A new test for direct effects of past exposures on a subsequent outcome is described.},
  author       = {Keogh, Ruth H and Daniel, Rhian M and VanderWeele, Tyler J and Vansteelandt, Stijn},
  issn         = {0002-9262},
  journal      = {AMERICAN JOURNAL OF EPIDEMIOLOGY},
  keywords     = {direct effect,indirect effect,inverse probability weight,longitudinal study,marginal structural model,sequential conditional mean model,time-varying confounder,total effect,MARGINAL STRUCTURAL MODELS,INVERSE PROBABILITY WEIGHTS,CAUSAL INFERENCE,G-COMPUTATION,ADJUSTMENT},
  language     = {eng},
  number       = {5},
  pages        = {1085--1092},
  title        = {Analysis of longitudinal studies with repeated outcome measures : adjusting for time-dependent confounding using conventional methods},
  url          = {http://dx.doi.org/10.1093/aje/kwx311},
  volume       = {187},
  year         = {2018},
}

Altmetric
View in Altmetric
Web of Science
Times cited: