Disclosure: David B. Allison has received book royalties, grants, consulting fees, and donations from multiple profit and nonprofit entities with interests in obesity, including pharmaceutical companies which compete with the manufacturers of sibutramine and from the manufacturers of sibutramine. The remaining authors declare no conflict of interest.
Article first published online: 26 MAR 2013
Copyright © 2012 The Obesity Society
Volume 21, Issue 2, pages 398–404, February 2013
How to Cite
Robertson, H. T., Campos, G. d. l. and Allison, D. B. (2013), Turning the analysis of obesity–mortality associations upside down: Modeling years of life lost through conditional distributions. Obesity, 21: 398–404. doi: 10.1002/oby.20019
Funding agencies: This research was sponsored by NIH grants T32HL079888, T32HL072757, P30DK056336, and R01DK076771.
- Issue published online: 26 MAR 2013
- Article first published online: 26 MAR 2013
- Accepted manuscript online: 28 AUG 2012 10:57AM EST
- Manuscript Accepted: 24 JUN 2012
- Manuscript Received: 6 OCT 2011
- NIH. Grant Numbers: T32HL079888, T32HL072757, P30DK056336, R01DK076771
We demonstrate the utility of parametric survival analysis. The analysis of longevity as a function of risk factors such as body mass index (BMI; kg/m2), activity levels, and dietary factors is a mainstay of obesity research. Modeling survival through hazard functions, relative risks, or odds of dying with methods such as Cox proportional hazards or logistic regression are the most common approaches and have many advantages. However, they also have disadvantages in terms of the ease of interpretability, especially for non-statisticians; the need for additional data to convert parameter estimates to estimates of years of life lost (YLL); debates about the appropriate time scale in the model; and an inability to estimate median survival time when the censoring rate is too high.
Design and Methods:
We will conduct parametric survival analyses with multiple distributions, including distributions that are known to be poor fits (Gaussian), as well as a newly discovered “Compressed Gaussian”'' distribution.
Parametric survival analysis models were able to accurately estimate median survival times in a population-based data set of 15,703 individuals, even for distributions that were not good fits and the censoring rate was high, due to the central limit theorem.
Parametric survival models are able to provide more direct answers, and in our analysis of an obesity-related data set, gave consistent YLL estimates regardless of the distribution used. We recommend increased consideration of parametric survival models in chronic disease and risk factor epidemiology.