CONFOUNDING BIAS:
Often shorted to Confounding and sometimes not classified as a bias. Generally, a distortion of an estimate of association between variables due to the intermixing of effects of other variables/charactersitics1. Confounding is a well-developed topic in medical science. Thus, there are several intricacies that need to be considered before drawing judgments on whether an association is confounded or not; which are beyond the scope of this dictionary.
Typically confounding is evaluated with respect the relationship(s) between exposure(s) and disease(s)/outcome(s). However, theoretically all relationships between any variables/characteristics may be confounded in a medical research study; which highlights the importance of using casual logic when planning a study.
It is debated as to whether Confounding should be classified as a bias. Confounding is thought to exist, to some degree, in all research studies, no matter what the researchers or participants do, as it is a component of natural, biological, and social diversity. Thus confounding cannot be completely eliminated or observed, as biological entities (animal, humans, plants etc) are infinitely complex. Because Confounding occurs due to natural diversity, it is questionable as to whether Confounding should be thought of as a ‘mistake’ or an ‘error’ in the process of research (see: Bias (Definition)).
Broadly, methods used to address Confounding include randomization, restriction of individuals/characteristics, stratification, matching, and statistical adjustment. The choice to use a method depends on context; as some methods may make Confounding worse, introduce new Confounding, or introduce other biases that did not previously exist. Confounding, and the methods used to address it, have many similarities with Selection Bias, thus understanding Selection Bias is important for understanding Confounding. Also see: Treatment Selection Bias, Covariate Bias, Omitted Variable Bias, Susceptibility Bias, Allocation Bias, Selection Bias, and Structural Selection Bias.
References:
1. Porta M, ed. A Dictionary of Epidemiology. Sixth ed. New York, NY: Oxford University Press 2014. (Link to Reference)