Reply to K. Jackson et al and W. Leppert

We thank Jackson et al1 and Leppert2 for the opportunity to

respond to their concerns regarding our article.3

Our main point of contention is the belief, on the part of Jackson

et al,1 that unblinded, nonrandomized, prospective audits can give a

true indication of the efficacy or toxicity of a drug. Open-label studies

do not meet the specific scientific definition of control. The gross

underestimation and overestimation of effects in uncontrolled trials,

especially studies that involve small numbers of participants, have

been well documented.4 This issue is particularly emphasized by the

high placebo response rate (27%) demonstrated in our blinded, randomized

controlled trial.