A measurement bias due to systematic differences between the compared groups in how the outcomes are ascertained.