|
Activity Number:
|
209
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #309589 |
|
Title:
|
Comparison of Modeling Approaches to Deriving Normative Values of Bone Mineral Density from NHANES III
|
|
Author(s):
|
Qing Wu*+ and John Lefante and Jeanette Magnus and Janet Rice
|
|
Companies:
|
Tulane University and Tulane University and Tulane University and Tulane University
|
|
Address:
|
, , ,
|
|
Keywords:
|
Bone Mineral Density ; NHANES III ; osteoporosis ; Model
|
|
Abstract:
|
Developing accurate normative reference bone mineral density (BMD) values is important to diagnose osteoporosis and prevent fractures. The study aim was to compare linear and polynomial regression models by residual analysis to identify the best model to derive normative BMD values at various skeletal sites using NHANES III. Gender- and race-specific group data analyses were conducted on 9779 aged 20--65 subjects who had no conditions or medication affecting bone metabolisms. Sampling and design effects were addressed using STATA 9.1. The results show that polynomial regression equations provide a significantly better fit than linear equations. These race- and gender- specific polynomial regression models, with their respective lower limits of normal, have the potential to improve diagnostic accuracy for low bone mass and can be use for populations other than postmenopausal women.
|
