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?

Klaus Abt, PhD
University of Frankfurt Medical School Frankfurt, Germany

JAMA. 1997;277(4):297.

	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 article by Dr Borhani et al¹ contains judgments of adverse event frequencies that call for some comments regarding the multiplicity^2,3 aspects of testing events (listed in their Table 4). Although the authors mention possible chance occurrences of significances, they suggest that the results are "of concern."

Concentrating on the 6 "major vascular events" (ie, stroke, myocardial infarction, sudden death, congestive heart failure, angina pectoris, and other cardiovascular disease deaths) the problem of multiple testing may be circumvented. Although the authors state that only 1 major event was counted per patient (leading to 6 mutually independent groups of patients each with 1 major event), the total number of events is 28 and not 25 (as indicated in their Table 4). Ignoring this deviation as well as the fact that the 6 test results are based on randomized patients, the 6 tests may be considered as essentially independent of . . . [Full Text PDF of this Article]

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