Konfundierer

Konfundierer is a term used primarily in statistics and epidemiology to denote a variable that can distort the apparent relationship between an exposure and an outcome. A Konfundierer is associated with both the exposure and the outcome and is not on the causal path from exposure to outcome. Such a variable can generate a spurious association or mask a true one, leading to confounding bias in observational studies.

Examples include age, socioeconomic status, or smoking, which may influence both the exposure (for example, lifestyle factors) and the outcome (such as disease risk). In studies of physical activity and cardiovascular disease, age and smoking are common Konfundierer that can bias the estimated effect if not properly accounted for.

Types and identification: Confounders can be measured (observed and adjustable in analysis) or unmeasured (not observed, leading to residual confounding). Causal diagrams, such as directed acyclic graphs, help hypothesize which variables may act as Konfundierer and how they relate to exposure and outcome. Researchers examine whether adjusting for a variable substantially changes the exposure–outcome estimate.

Control and mitigation: Methods to control Konfundierer include study design choices (randomization, restriction, matching) and statistical approaches (multivariable regression, propensity score methods, instrumental variables). It is important not to adjust for mediators or colliders, which can introduce new biases.

Limitations: Even with careful adjustment, residual confounding may persist, especially when confounders are unmeasured or poorly measured. Consequently, causal conclusions from observational data should be drawn with caution. See also confounding, causal inference, and propensity scores.