Advanced search
1 file | 211.00 KB

Estimation of controlled direct effects on a dichotomous outcome using logistic structural direct effect models

(2010) BIOMETRIKA. 97(4). p.921-934
Author
Organization
Abstract
We consider the problem of assessing whether an exposure affects a dichotomous outcome other than by modifying a given mediator. The standard approach, logistic regression adjusting for both exposure and the mediator, is known to be biased in the presence of confounders for the mediator-outcome relationship. Because additional regression adjustment for such confounders is only justified when they are not affected by the exposure, inverse probability weighting has been advocated, but is not ideally tailored to mediators that are continuous or have strong measured predictors. We overcome this limitation by developing inference for a novel class of causal models that are closely related to Robins' logistic structural direct effect models, but do not inherit their difficulties of estimation. We study identification and efficient estimation under the assumption that all confounders for the exposure-outcome and mediator-outcome relationships have been measured, and find adequate performance in simulation studies. We discuss extensions to case-control studies and relevant implications for the generic problem of adjustment for time-varying confounding.
Keywords
MEAN MODELS, CAUSAL INFERENCE, POTENTIAL OUTCOMES, RANDOMIZED-TRIALS, MEDIATION, INTERVENTIONS, Case-control study, Direct effect, G-estimation, Instability, Logistic structural mean model, Mediation, Pathway, Time-varying confounding

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 211.00 KB

Citation

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

Chicago
Vansteelandt, Stijn. 2010. “Estimation of Controlled Direct Effects on a Dichotomous Outcome Using Logistic Structural Direct Effect Models.” Biometrika 97 (4): 921–934.
APA
Vansteelandt, S. (2010). Estimation of controlled direct effects on a dichotomous outcome using logistic structural direct effect models. BIOMETRIKA, 97(4), 921–934.
Vancouver
1.
Vansteelandt S. Estimation of controlled direct effects on a dichotomous outcome using logistic structural direct effect models. BIOMETRIKA. 2010;97(4):921–34.
MLA
Vansteelandt, Stijn. “Estimation of Controlled Direct Effects on a Dichotomous Outcome Using Logistic Structural Direct Effect Models.” BIOMETRIKA 97.4 (2010): 921–934. Print.
@article{1234543,
  abstract     = {We consider the problem of assessing whether an exposure affects a dichotomous outcome other than by modifying a given mediator. The standard approach, logistic regression adjusting for both exposure and the mediator, is known to be biased in the presence of confounders for the mediator-outcome relationship. Because additional regression adjustment for such confounders is only justified when they are not affected by the exposure, inverse probability weighting has been advocated, but is not ideally tailored to mediators that are continuous or have strong measured predictors. We overcome this limitation by developing inference for a novel class of causal models that are closely related to Robins' logistic structural direct effect models, but do not inherit their difficulties of estimation. We study identification and efficient estimation under the assumption that all confounders for the exposure-outcome and mediator-outcome relationships have been measured, and find adequate performance in simulation studies. We discuss extensions to case-control studies and relevant implications for the generic problem of adjustment for time-varying confounding.},
  author       = {Vansteelandt, Stijn},
  issn         = {0006-3444},
  journal      = {BIOMETRIKA},
  language     = {eng},
  number       = {4},
  pages        = {921--934},
  title        = {Estimation of controlled direct effects on a dichotomous outcome using logistic structural direct effect models},
  url          = {http://dx.doi.org/10.1093/biomet/asq053},
  volume       = {97},
  year         = {2010},
}

Altmetric
View in Altmetric
Web of Science
Times cited: