Deconfounding confounding part 2: using directed acyclic graphs (DAGs)

Attia, John R; Oldmeadow, Christopher; Holliday, Elizabeth G; Jones, Michael P

doi:10.5694/mja16.01167

Topics

Statistics, epidemiology and research design

In the previous article on confounding in this series,1 we presented the traditional explanation of a confounder. Over the past few decades, it has become clear that this definition has many limitations. For example, confounding can be induced by a network of variables rather than just a single variable, and adjusting for potential confounders can paradoxically increase confounding.

1 Centre for Clinical Epidemiology and Biostatistics, University of Newcastle, Newcastle, NSW
2 John Hunter Hospital, Newcastle, NSW
3 Hunter Medical Research Institute, Newcastle, NSW
4 Macquarie University, Sydney, NSW

Correspondence: john.attia@newcastle.edu.au

Competing interests:

No relevant disclosures.

1. Attia JR, Jones MP, Hure A. Deconfounding confounding part I: traditional explanations. Med J Aust 2017; 206: 244-245. <MJA full text>
2. Hernán MA, Robins JM. Causal inference. Boca Raton, FL: Chapman & Hall/CRC, 2016.
3. Glymour MM, Greenland S. Causal diagrams. In: Rothman KJ, Greenland S, Lash TL, editors. Modern epidemiology, 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2008; 183-209.
4. Pearl J. Causality, 2nd ed. New York: Cambridge University Press, 2009.
5. Shrier I, Platt RW. Reducing bias through directed acyclic graphs. BMC Med Res Methodol 2008; 8: 70.
6. Schisterman EF, Cole SR, Platt RW. Overadjustment bias and unnecessary adjustment in epidemiologic studies. Epidemiology 2009; 20: 488-495.
7. Textor J, Hardt J, Knüppel S. DAGitty: a graphical tool for analyzing causal diagrams. Epidemiology 2011; 22: 745.
8. Greenland S, Pearce N. Statistical foundations for model-based adjustments. Annu Rev Public Health 2015; 36: 89-108.

Online responses are no longer available. Please refer to our instructions for authors page for more information.