Evidence based medicine distinguishes observational studies from randomised controlled studies. Randomised controlled trial is a design that can provide the most compelling evidence that the study treatment causes the expected effect in patients. Randomised: each patient is assaigned to receive either treatment or placebo.Blind: the patients dont know which treatment they recive.If double-blinded, the researcher also doesnt know which treatment is being administered to any subject. "Blinding" prevents bias, and in "double-dummy" design all patients may get both placebo and active treatment in alternating periods of time.
Phase 0: first-in-human trials, enable go/no-go decisions to be based on humans and not animal models
Phase I: assess safety, tolerability and pharmacokinetic/dynamics of drug
Phase II: assess how well drug work and safety
Phase III: randomised, controlled multicentre on many patients, tests how well it works compared to current treatments.