Randomization of Human Subjects

It may be accepted as a maxim that a poorly or improperly designed study involving human subjects — one that could not possibly yield scientific facts (that is, reproducible observations) relevant to the question under study — is by definition unethical.1 If one accepts this maxim one must also accept that the design and analysis of clinical investigations should be as effective as possible either in reducing the number of people at risk or in increasing the strength of the conclusions. Dr. Weinstein (page 1278) discusses many of the technical issues involved in the use of decision theory and statistics . . .