Evidence-appraisal glossary
Confounding by indication
Confounding by indication is a bias in observational studies where the clinical reason a treatment was given is itself linked to the outcome. The apparent effect of the drug is mixed up with the effect of the condition that prompted it.
Also called: indication bias, confounding by severity.
Confounding by indication occurs because treatments are not assigned at random; sicker patients, or those with specific risk factors, are the ones who receive certain drugs, so their prognosis differs from the start. This can make an effective treatment look useless or even harmful if the underlying severity is not fully accounted for. It is closely related to channeling bias, and while methods like propensity scores can reduce it, they cannot adjust for severity that was never measured.
This is a plain-language methodology definition for reading research. It is general education, not medical advice.