Evidence-appraisal glossary
Confounding
Confounding occurs when a third factor is linked to both the exposure and the outcome, creating a misleading association. It can make an exposure look harmful or protective when the real driver is something else the two groups differ on, such as age or smoking.
Also called: confounder, confounding variable, lurking variable.
A confounder is a variable associated with both the exposure and the outcome, and not simply a step on the causal pathway between them. It can distort an observed association, making an exposure appear to cause an effect it does not, or masking a real one. The classic illustration: coffee drinking looks linked to lung cancer, but coffee drinkers were also more likely to smoke, and smoking causes the cancer. Here smoking confounds the coffee-cancer relationship. Researchers handle confounding by design (randomization, restriction, matching) or in analysis (stratification, multivariable adjustment, propensity scores). When reading a study, ask which confounders the authors measured and adjusted for, and whether important ones were left out. Residual confounding from unmeasured or poorly measured factors is a leading reason observational findings later fail to hold up. The question it lets you ask: could the reported association be explained by some other characteristic on which the compared groups differ?
This is a plain-language methodology definition for reading research. It is general education, not medical advice.