A method of comparing the results between two or more groups of patients intentionally subjected to different methods of treatment – or sometimes of prevention. Those entering the trial have to give their informed permission. They are allocated to their respective groups at random one group (controls) receives either PLACEBO or a traditional treatment, while the active group receives the treatment under test. Preferably, neither the subject nor the assessor should know which group a subject is in: this is known as a double-blind trial. A monitoring group, separate from the researchers, reviews results from time to time and has the power to stop the trial if one treatment is clearly much better than the other (or more dangerous). Once the trial is complete, the blinding codes are broken and researchers apply statistical techniques to determine the results.