Log in with your email address username.

×

Important notice

doctorportal Learning is on the move as we will be launching a new website very shortly. If you would like to sign up to dp Learning now to register for CPD learning or to use our CPD tracker, please email support@doctorportal.com.au so we can assist you. If you are already signed up to doctorportal Learning, your login will work in the new site so you can continue to enrol for learning, complete an online module, or access your CPD tracker report.

To access and/or sign up for other resources such as Jobs Board, Bookshop or InSight+, please go to www.mja.com.au, or click the relevant menu item and you will be redirected.

All other doctorportal services, such as Find A Doctor, are no longer available.

Deconfounding confounding part 1: traditional explanations

- Featured Image

The first article of this series1 presented a framework to assist in judging the presence of bias:

  • selection bias, or systematic error in how participants are identified or selected;

  • measurement bias, or systematic error in how variables are measured; and

  • analytical bias — also known as confounding — or systematic error in the measure of association or conclusion about causation, due to improper or incomplete analysis.

Selection and measurement bias should be managed pre-emptively by good design before the start of the study, but can be detected post hoc by critical appraisal. No statistical method removes the effect of selection or measurement bias post hoc, although there are methods that allow us to model different degrees of bias and evaluate the effect on the measure of association.1 Confounding is slightly different in that it can be adjusted for in the analysis, as long as its sources are understood and measured without too much error.

What is confounding?

A confounder has been traditionally defined as a variable associated with both the exposure and outcome of interest without being an intermediate on the causal pathway between them, which causes a spurious or distorted estimate of the exposure–outcome association. This may be conceptually difficult…

email