|
Activity Number:
|
405
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #301589 |
|
Title:
|
On Interval Estimations for a Dichotomized Variable Derived from a Continuous Measure
|
|
Author(s):
|
Jessica L. Kim*+ and Jean Wang
|
|
Companies:
|
U.S. Food and Drug Administration and U.S. Food and Drug Administration
|
|
Address:
|
1401 Rockville Pike, Rockville, MD, 20852,
|
|
Keywords:
|
dichotomized variable ; tolerance limit ; data simulation ; RBC (Red Blood Cell) products ; percent of coverage
|
|
Abstract:
|
When an analysis is done with a dichotomized variable from a continuous measurement, the result can be interpreted differently. In this paper, tolerance limits with various coverage levels using the original data scale will be estimated. Then, the results of tolerance intervals will be used as a threshold value to dichotomize the variable. The corresponding tolerance limits of the dichotomized variable will be investigated and compared to the results of the original scale. Data from simulation as well as in vivo studies of Red Blood Cell products will be discussed and commented as an application. An interval that predicts p percent of coverage for a dichotomized variable derived from a continuous measure will be investigated.
|
