Evidence-based medicine. Part 2: Randomised controlled trials
The modern randomised controlled trial is the most reliable inferential means of investigating treatment effects. Its development during the 20th century turned clinical medicine into a rigorous experimental science. The second in a series, this article discusses the practical challenges of conducting trials including randomisation, allocation concealment and blinding, and explains concepts such as intention to treat and equipoise. Large diabetes randomised controlled trials are used as examples.