Residual Confounding

In statistics, a confounding variable (also confounding factor, lurking variable, a confound, or confounder) is an extraneous variable in a statistical model that correlates (positively or negatively) with both the dependent variable and the independent variable. The methodologies of scientific studies therefore need to control for these factors to avoid a false positive (Type I) error; an erroneous conclusion that the dependent variables are in a causal relationship with the independent variable. Such a relation between two observed variables is termed a spurious relationship. Thus, confounding is a major threat to the validity of inferences made about cause and effect, i.e. internal validity, as the observed effects should be attributed to the independent variable rather than the confounder.

公卫百科

Residual confounding refers to confounding that has been incompletely controlled, so that confounding effects of some factors may remain in the observed treatment-outcome effect. Residual confounding is often only discussed qualitatively without trying to quantify its effect. Yet methods are available to attempt to assess the magnitude of residual confounding after adjusted effects have been obtained. (Psaty 1999; Lash and Fink, 2003).

RESIDUAL CONFOUNDING SHOULD BE ASSESSED AND APPROACHES TO ESTIMATING ITS EFFECT, INCLUDING SENSITIVITY ANALYSES, SHOULD BE INCLUDED. 公卫百科