This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
To the Editor.—
Cooper's answer to the comment by Oseasohn in the LETTERS section (236:1449, 1976) nicely illustrates the corner into which we have painted ourselves: We are so impressed by less than adequate studies that we foreclose the opportunity to do an adequate study. Oseasohn states that we do not have "properly designed trials" to support the Cooper conclusion. Cooper seems to miss the point when he implies that a "prospective study" would satisfy the requirement of being "properly designed."Feinstein (a member of the Editorial Board of The Journal), Spodick, and others have repeatedly pointed out that this kind of difference is not just a difference of opinion. The difference relates to the quality of the data required before the data are considered adequate on which to base an opinion. Well-known names can be found on both sides of this issue, just as every other. It is not
Gilson JS. Properly Designed Trials. JAMA. 1977;237(1):27. doi:10.1001/jama.1977.03270280029014