Double-blind

A double-blind trial is one in which neither participants nor investigators (including outcome assessors and, ideally, sponsors performing analyses) know treatment assignments until the database is locked. Double-blinding—paired with randomization and a credible control—is widely regarded as the methodologic gold standard for minimizing performance and detection bias in clinical research. Cannabis trials face well-documented challenges to maintaining the blind: THC's psychoactive and physiologic effects (euphoria, tachycardia, conjunctival injection, dry mouth) are readily recognized by experienced users and by clinicians conducting assessments, and the characteristic odor of inhaled cannabis complicates staff blinding. The 2017 NASEM report's methodology chapter explicitly discusses these issues and recommends post-hoc assessment of blinding integrity, use of active placebos, and consideration of dose-ranging designs to mitigate functional unblinding. Trials of purified cannabidiol generally maintain the blind more successfully. → See also: Placebo-controlled, IRB.