To the Editor.
—The discussion by Dr Eddy1 of what is needed for evidence to be "sufficient" when evaluating medical treatments deserves further exploration. Even if we can agree on appropriate end points, study designs, and use of expert knowledge, the question of how certain we must be that treatment A is beneficial compared with treatment B is still
Wears RL. What Is Necessary for Proof? Is 95% Sure Unrealistic?. JAMA. 1994;271(4):272. doi:10.1001/jama.1994.03510280034022
Customize your JAMA Network experience by selecting one or more topics from the list below.