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:—
Dr. Schor's letter is an excellent example for elementary medical courses to show how one can take a set of data which demonstrate a probable drug effect for a particular class of patients, and yet by combining patients with different disease states, conclude that the drug had no effect at all. He says, "The statistical test applied to each of the 21 patients shown in the Table demonstrates that in some, cholesterol levels went down, and in others they did not. That is all." This completely ignores the fact that a highly significant decrease in serum cholesterol level occurred in ten of 14 patients with one form of hyperlipoproteinemia (type II), but in none of the patients with other lipid disorders. Dr. Schor's doubt about the reproducibility of electrophoresis patterns does not reflect the experience of most investigators in this field.It is true that ideally a
Fallon HJ, Woods JW, Elston RC. Hyperlipoproteinemia— A Difference of Opinion. JAMA. 1968;206(2):376. doi:10.1001/jama.1968.03150020091029