Randomised trial
A trial in which a coin toss (done by computer) decides which treatment each patient gets, so the groups are alike in every way except the treatment. It is the only reliable way to prove a treatment causes a benefit.
Without randomisation, patients who receive a new drug tend to differ from those who do not (fitter, younger, treated at better centres), and those differences, not the drug, may explain a better outcome; randomising removes this bias, and analysing every patient in the group they were assigned to (intention-to-treat) preserves it. The control group receives the current standard of care, sometimes with a placebo added so that the two arms look identical, and the trial is powered so that a real difference of a given size is unlikely to be missed. Large randomised trials are slow and expensive, which is why surrogate endpoints, single-arm accelerated approvals and real-world evidence are all attempts to shortcut them, each with known pitfalls.