A randomized controlled trial (RCT) is the most definitive tool for ev
aluation of the effectiveness of an intervention and can establish a c
ause-and-effect relationship between an intervention and an improved d
isease outcome. However, the undertaking of an RCT does not guarantee
valid results, and the findings of RCTs of the same intervention are o
ften discrepant. While many reported trials are of high quality, a sig
nificant number have deficiencies in design, conduct, analysis, or int
erpretation of results. Medical practitioners must become familiar wit
h the methodology of RCTs in order to assess the validity of the repor
ted findings and the relevance of the results to their own patients. T
rialists must report sufficient information about the methods used se
that readers can judge for themselves if the trial worked as planned.
(C) 1997 Wiley-Liss, Inc.