Printer Friendly

Statistical vs. clinical significant numbers.

Iread with great interest the "Tips from clinical experts Q & A" in the April 2003 issue of MLO, page 39, "Comparing two analyzers," particularly the answer in testing the null hypothesis. The author states that the traditional cut-off value indicating significance is p<0.001. According to the Michelin Guide scale p<0.05 is significant, p<0.01 is highly significant and p<0.001 is extremely significant. I think that needs to be clarified.

Just because something is statistically significant does not mean that it is clinically significant. If the laboratory only used a narrow range of specimens, such as in a normal range study on both instruments, a small difference can make the comparison have a value of p<0.001 but the clinical significance is miniscule. (See McGlasson DL, et al., Rapid removal of platelets from plasma utilizing the Hepcheck heparin removal filter, Blood Coagulation and Fibrinolysis 8, 16-20, 1997).

[ILLUSTRATION OMITTED]

A very well-explained definition of what that value means is the explanation about the probability indicating that the p<0.001, due to chance alone, is less than one in a thousand. Sometimes, laboratorians get caught up in numbers that do not have a clinically significant relevance in relation to their statistical outcome.

--David L. McGlasson, MS, CLS/NCA

Research Scientist

59th Clinical Research Squadron

Lackland AFB, TX

MLO welcomes letters to the editor. We ask that you include a phone number for verification. While we prefer to publish the writer's name, we will publish a letter with "name withheld by request," but our editorial staff must have the writer's name confirmed for our files. MLO reserves the right to edit any letter for style and length.
COPYRIGHT 2003 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Readers Respond
Author:McGlasson, David L.
Publication:Medical Laboratory Observer
Article Type:Letter to the Editor
Date:Dec 1, 2003
Words:283
Previous Article:Summer in the dead of winter.
Next Article:Multiple views misleading.


Related Articles
Why readers respond. (Feedback).
Another candidate for the "top 10".
Testing toxic compounds in human subjects: ethical standards and good science.
Uniform Reporting Manual, 4th Edition.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters