confounding variable

Syllable breakdown

Dictionary meaning

a variable in an epidemiological or statistical study that is associated with both the exposure of interest and the outcome, creating a spurious or distorted apparent relationship between them; confounding occurs when the observed association between an exposure and outcome is partly or entirely attributable to a third factor that tracks with both; for example, a study finding that coffee drinking is associated with lung cancer may be confounded by smoking if smokers are more likely to drink coffee; confounding can be controlled in study design through randomisation, restriction, or matching, and in analysis through stratification, multivariable regression, or propensity score methods; the identification and appropriate handling of confounders is central to the valid interpretation of observational research and the distinction between association and causation