View on Pubmed.gov 21627630 | View on doi.org 10.1111/j.1541-...
A new criterion for confounder selection.
- Departments of Epidemiology and Biostatistics, Harvard School of Public Health, 677 Huntington Avenue, Boston, Massachusetts 02115, USA. tvanderw@hsph.harvard.edu
Abstract
We propose a new criterion for confounder selection when the underlying causal structure is unknown and only limited knowledge is available. We assume all covariates being considered are pretreatment variables and that for each covariate it is known (i) whether the covariate is a cause of treatment, and (ii) whether the covariate is a cause of the outcome. The causal relationships the covariates have with one another is assumed unknown. We propose that control be made for any covariate that is either a cause of treatment or of the outcome or both. We show that irrespective of the actual underlying causal structure, if any subset of the observed covariates suffices to control for confounding then the set of covariates chosen by our criterion will also suffice. We show that other, commonly used, criteria for confounding control do not have this property. We use formal theory concerning causal diagrams to prove our result but the application of the result does not rely on familiarity with causal diagrams. An investigator simply need ask, "Is the covariate a cause of the treatment?" and "Is the covariate a cause of the outcome?" If the answer to either question is "yes" then the covariate is included for confounder control. We discuss some additional covariate selection results that preserve unconfoundedness and that may be of interest when used with our criterion.
References
Neuropsychol Rev. 2008 Sep;18(3):194-213
J Clin Epidemiol. 1993 Jan;46(1):57-63
Stat Med. 2009 Apr 30;28(9):1416-20; author reply 1420-3
Stat Med. 2007 Jan 15;26(1):20-36
Cited by 377 citations (total)
Journal Article Research Support, Non-U.S. Gov't
Augmented backward elimination: a pragmatic and purposeful way to develop statistical models.
Electronic publication date December 17, 2015
Daniela Dunkler [...] Georg Heinze et al.
Statistical models are simple mathematical rules derived from empirical data describing the association between an outcome and several explanatory variables....
Journal Article
The ACCE method: an approach for obtaining quantitative or qualitative estimates of residual confounding that includes unmeasured confounding.
Electronic publication date January 13, 2015
Eric G Smith
BACKGROUND
Nonrandomized studies typically cannot account for confounding from unmeasured factors.
METHOD
A method is presented that exploits the recently-identified phenomenon of  "confounding...
Conflict of interest statement
MeSH
Specialties
- •
Copyright
© 2011, The International Biometric Society.