Confounding: what it is and how to deal with it

Abstract

As confounding obscures the 'real' effect of an exposure on outcome, investigators performing etiological studies do their utmost best to prevent or control confounding. Unfortunately, in this process, errors are frequently made. This paper explains that to be a potential confounder, a variable needs to satisfy all three of the following criteria: (1) it must have an association with the disease, that is, it should be a risk factor for the disease; (2) it must be associated with the exposure, that is, it must be unequally distributed between exposure groups; and (3) it must not be an effect of the exposure; this also means that it may not be part of the causal pathway. In addition, a number of different techniques are described that may be applied to prevent or control for confounding: randomization, restriction, matching, and stratification. Finally, a number of examples outline commonly made errors, most of which result from 'overadjustment' for variables that do not satisfy the criteria for potential confounders. Such an example of an error frequently occurring in the literature is the incorrect adjustment for blood pressure while studying the relationship between body mass index and the development of end-stage renal disease. Such errors will introduce new bias instead of preventing it.