|
Activity Number:
|
393
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #307879 |
|
Title:
|
Intentional Weight-Loss Effects on Mortality Rate Modeled as a Latent Variable Problem
|
|
Author(s):
|
Gary L. Gadbury*+
|
|
Companies:
|
University of Missouri-Rolla
|
|
Address:
|
Mathematics and Statistics, Rolla, MO, 65409,
|
|
Keywords:
|
causation ; nonestimable ; obesity ; potential outcomes
|
|
Abstract:
|
Obesity is associated with increased mortality rate and short-term weight loss improves risk factors for mortality. It has not been convincingly shown that weight loss among obese persons results in reduced mortality rate. Studies have pointed out that weight loss is often a sign of illness and intentional weight loss (IWL) should be separated from unintentional weight loss (UWL). Coffey et al. (Stats in Med, 2005) considered the problem of disentangling the effects of IWL on mortality from UWL. They showed that the effect of IWL on mortality is a nonestimable parameter. A potential outcomes framework is presented to illuminate required assumptions and for defining estimable bounds for the nonestimable parameter. The framework is extended to the scenario where subjects change their intention to lose weight over two time periods. A Bayesian approach to this problem is also discussed.
|
