Activity Number:
|
286
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:20 PM
|
Sponsor:
|
ENAR
|
Abstract - #309741 |
Title:
|
Bayesian Analysis of the Effect of Intentional Weight Loss on Mortality Rate
|
Author(s):
|
Nengjun Yi and Shouluan Ding and Scott W. Keith*+ and Christopher S. Coffey and David B. Allison
|
Companies:
|
The University of Alabama at Birmingham and The University of Alabama at Birmingham and The University of Alabama at Birmingham and The University of Alabama at Birmingham and The University of Alabama at Birmingham
|
Address:
|
Department of Biostatistics, Birmingham, AL, 35294,
|
Keywords:
|
Bayesian analysis ; latent variables ; mortality ; obesity ; weight loss
|
Abstract:
|
The effect of weight loss (WL) on mortality rate is widely studied in obesity research. Separating the effects of intentional weight loss (IWL) from unintentional weight loss (UWL) continues to be a challenge. It has been shown that WL among people intending to lose weight is not equivalent to IWL. We constructed Bayesian latent variable linear models that allow the separation of IWL and UWL effects among those intending to lose weight by augmenting their unobserved UWL with information from observed WL among those not intending to lose weight. This approach provides estimates of IWL and UWL effects as well as any other parameters of interest. We applied our method to a real rodent caloric restriction study dataset. Our results suggest that IWL has a substantial beneficial effect on mouse lifespan, in contrast to UWL. We also discuss extensions to human data and censored outcomes.
|