This Article  • References  • Full text PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Contact me when this article is cited  Related Content  • Similar articles in JAMA  Social Bookmarking   What's this?

Stephen Warshafsky, MD
New York Medical College Valhalla

JAMA. 1991;266(8):1079-1080.

	Since this article does not have an abstract, we have provided the first 150 words of the full text PDF and any section headings.

To the Editor.

—The recent article by Davidson et al¹ claims a power of 0.90 with an of.05 to detect a 0.39-mmol/L (15-mg/dL) change in total cholesterol and low-density lipoprotein cholesterol (LDL-C) levels between the control and experimental groups. In my estimation, their experimental groups of 20 to 23 subjects each and placebo group of 15 are not sufficient to achieve such a power.

As serum cholesterol levels are considered a continuous variable, an estimate for population size for each experimental and placebo group can be calculated by using the Student t test to compare means between placebo and each experimental group, using the following formula: N = ([1/q₁ + 1/q₂]S²[z + z_β]²)/E², where N = total number of subjects required, S=SD from the mean, E = effect size, z = the standard normal deviate for (z = 1.96 when =.05), z . . . [Full Text PDF of this Article]

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter     What's this?