Refers to bias arising from differential assessment of outcomes, whether by patients or investigators, influenced by their knowledge of the assignment of patients to intervention or control groups. Blinding of patients and investigators to treatment assignment is a technique used to manage detection bias.