THE RANDOMIZED controlled clinical trial (RCT) has become the accepted standard for demonstrating the true therapeutic effects in evaluating most forms of treatment. Its scientific and ethical bases are well established. Yet it is standard practice, at least for trials of new pills and injections, to take certain decisions out of physicians' hands by "blinding" and by randomization, letting chance rather than prejudice decide which patients get the trial treatment. This is partly because physicians' behavior has consistently been heavily biased in relation to therapeutic trials. A flagrant expression is found in the schizoid behavior of journal reviewers and institutional committees that insist on high standards of proof for pills and low standards for surgical procedures. Indeed, an important reason for the existence of the Food and Drug Administration is to ensure optimum scientific and ethical behavior in assessing the efficacy and safety of new medicines. (There is no FDA