Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Mar;34(3):211-219.
doi: 10.1007/s10654-019-00494-6. Epub 2019 Mar 6.

Principles of confounder selection

Tyler J VanderWeele  1
Affiliations

Principles of confounder selection

Tyler J VanderWeele. Eur J Epidemiol. 2019 Mar.

Abstract

Selecting an appropriate set of confounders for which to control is critical for reliable causal inference. Recent theoretical and methodological developments have helped clarify a number of principles of confounder selection. When complete knowledge of a causal diagram relating all covariates to each other is available, graphical rules can be used to make decisions about covariate control. Unfortunately, such complete knowledge is often unavailable. This paper puts forward a practical approach to confounder selection decisions when the somewhat less stringent assumption is made that knowledge is available for each covariate whether it is a cause of the exposure, and whether it is a cause of the outcome. Based on recent theoretically justified developments in the causal inference literature, the following proposal is made for covariate control decisions: control for each covariate that is a cause of the exposure, or of the outcome, or of both; exclude from this set any variable known to be an instrumental variable; and include as a covariate any proxy for an unmeasured variable that is a common cause of both the exposure and the outcome. Various principles of confounder selection are then further related to statistical covariate selection methods.

Keywords: Causal inference; Collider; Confounder; Covariate adjustment; Selection.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
Confounding by covariates C of the relationship between exposure A and outcome Y
Fig. 2
Fig. 2
Controlling for pre-exposure covariate L introduces bias in the relationship between exposure A and outcome Y because L is a collider on the path from A to Y, since it is a common effect of U1 and U2
Fig. 3
Fig. 3
Controlling for measured covariate C, even in the presence of unmeasured variable U, eliminates confounding of the relationship between exposure A and outcome Y, even though C itself is not a common cause of A and Y
Fig. 4
Fig. 4
In the presence of uncontrolled confounding between exposure A and outcome Y induced by unmeasured variable U, controlling for the instrument Z can amplify the bias induced by U
Fig. 5
Fig. 5
Control for a proxy confounder C1 of the true unmeasured confounder U will often, but not always, reduce confounding bias in the relationship between exposure A and outcome Y
See this image and copyright information in PMC

Comment in

References

    1. Pearl J. Causal diagrams for empirical research (with discussion) Biometrika. 1995;82:669–710.
    1. Pearl J. Causality: models, reasoning, and inference. 2. Cambridge: Cambridge University Press; 2009.
    1. Huang Y, Valtorta M. Pearl’s calculus of interventions is complete. In: Twenty second conference on uncertainty in artificial intelligence.
    1. Shpitser I, VanderWeele TJ, Robins JM. On the validity of covariate adjustment for estimating causal effects. In: Proceedings of the 26th conference on uncertainty and artificial intelligence. Corvallis: AUAI Press; (2010), p. 527–536.
    1. Greenland S. Quantifying biases in causal models: classical confounding vs collider-stratification bias. Epidemiology. 2003;14:300–306. - PubMed

LinkOut - more resources