Research Methods

Confounding

A variable that is associated with both the exposure and the outcome, potentially distorting the apparent relationship between them.

In Depth

Confounding is a major challenge in observational cannabis research. Cannabis users differ from non-users in many ways (tobacco use, alcohol use, socioeconomic status, mental health) that independently affect health outcomes. For example, the association between cannabis use and lung cancer is confounded by tobacco co-use. Researchers use statistical methods (multivariable regression, propensity score matching) to control for known confounders, but unmeasured confounding always remains a concern in observational studies.