|
Activity Number:
|
361
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #304178 |
|
Title:
|
When is a Diagnostic Test Informative?
|
|
Author(s):
|
Lakshmi Vishnuvajjala*+
|
|
Companies:
|
FDA
|
|
Address:
|
1350 Piccard Drive, HFZ-550, Rockville, MD, 20850,
|
|
Keywords:
|
Diagnostic ; Test ; Informative ; Prevalence ; False positive ; False negative
|
|
Abstract:
|
For a diagnostic test to be informative, it needs to identify a greater proportion of patients with the condition as positive compared to the proportion of patients without the condition. All the performance measures deal with when a test is informative statistically. Does all this necessarily translate into being informative in the clinical sense? For a test with 99% sensitivity and 61% specificity, the sum is 160%; it is also 160% for a test with 80% sensitivity and 80% specificity. Are both the tests equally informative? What role does prevalence play? Ultimately, the usefulness of the test depends on the consequences of false positives and false negatives. These issues will be discussed with illustrative examples.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
