Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The UGDP Mortality Results

The UGDP Mortality Results To the Editor.— The article by Cornfield analyzing the University Group Diabetes Program (UGDP) mortality findings (217:1676, 1971) raises several points concerning the role of the physician vis-a-vis the statistician in evaluating clinical research. A major part of the article was devoted to investigating the unlikely possibility of a breakdown in the randomization scheme and, of wider import, to analyzing whether "the luck of the draw" using randomization provided groups similar enough in baseline charateristics that the outcomes of the different treatment groups could be compared. "To see how bad the luck of the draw really was," Cornfield notes that none of the baseline characteristics were significantly different among treatment groups at the P =.05 level. However, this is a medical question and the problem of whether groups are similar enough for treatment results to be compared cannot be resolved using a statistical test of significance. If, for example, http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JAMA American Medical Association

The UGDP Mortality Results

Shapiro, Alan R.
JAMA , Volume 219 (13) – Mar 27, 1972
Download PDF
2 pages

Loading next page...
 
/lp/american-medical-association/the-ugdp-mortality-results-GXwnlyKJie

References (2)

Publisher
American Medical Association
Copyright
Copyright © 1972 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
ISSN
0098-7484
eISSN
1538-3598
DOI
10.1001/jama.1972.03190390045019
Publisher site
See Article on Publisher Site

Abstract

To the Editor.— The article by Cornfield analyzing the University Group Diabetes Program (UGDP) mortality findings (217:1676, 1971) raises several points concerning the role of the physician vis-a-vis the statistician in evaluating clinical research. A major part of the article was devoted to investigating the unlikely possibility of a breakdown in the randomization scheme and, of wider import, to analyzing whether "the luck of the draw" using randomization provided groups similar enough in baseline charateristics that the outcomes of the different treatment groups could be compared. "To see how bad the luck of the draw really was," Cornfield notes that none of the baseline characteristics were significantly different among treatment groups at the P =.05 level. However, this is a medical question and the problem of whether groups are similar enough for treatment results to be compared cannot be resolved using a statistical test of significance. If, for example,

Journal

JAMA – American Medical Association

Published: Mar 27, 1972

There are no references for this article.