DAVID D. CLARKE, M.D.
To the editor: The nomograms described by Young and associates (1) enable rapid estimation of sample size appropriate for study of an hypothesis. However, in evaluating completed negative clinical studies Feinstein (2) uses a more appropriate formula for dichotomous variables to calculate a precise value for the probability of a type 2 error.
Young and associates describe a study with two groups of 150 subjects each, one given drug A and the other drug B. The first group has a 40% cure rate and the second a 45% cure rate. The null hypothesis is accepted but a concern is raised:
CLARKE DD. Sample-Size Nomograms. Ann Intern Med. ;99:875–876. doi: 10.7326/0003-4819-99-6-875
Download citation file:
Published: Ann Intern Med. 1983;99(6):875-876.
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use