Morgan E Levine1, Eileen M Crimmins. 1. Davis School of Gerontology, University of Southern California, Los Angeles, California, 90089-0191.
Abstract
OBJECTIVES: Concepts such as Allostatic Load, Framingham Risk Score, and Biological Age were developed to combine information from multiple measures into a single latent variable that can be used to quantify a person's biological state. Given these varying approaches, the goal of this article is to compare how well these three measures predict subsequent all-cause and disease-specific mortality within a large nationally representative U.S. sample. METHODS: Our study population consisted of 9,942 adults, ages 30 and above from National Health and Nutrition Examination Survey III. Receiver Operating Characteristic curves and Cox Proportional Hazard models for the whole sample and for stratified age groups were used to compare how well Allostatic Load, Framingham Risk Score, and Biological Age predict ten-year all-cause and disease-specific mortality in the sample, for whom there were 1,076 deaths over 96,420 person years of exposure. RESULTS: Overall, Biological Age predicted 10-year mortality more accurately than other measures for the full age range, as well as for participants ages 50 to 69 and 70+. Additionally, out of the three measures, Biological Age had the strongest association with all-cause and cancer mortality, while the Framingham Risk Score had the strongest association with CVD mortality. CONCLUSIONS: Methods for quantifying biological risk provide important approaches to improving our understanding of the causes and consequences of changes in physiological function and dysregulation. Biological Age offers an alternative and, in some cases a more accurate summary approach to the traditionally used methods, such as Allostatic Load and Framingham Risk Score.
OBJECTIVES: Concepts such as Allostatic Load, Framingham Risk Score, and Biological Age were developed to combine information from multiple measures into a single latent variable that can be used to quantify a person's biological state. Given these varying approaches, the goal of this article is to compare how well these three measures predict subsequent all-cause and disease-specific mortality within a large nationally representative U.S. sample. METHODS: Our study population consisted of 9,942 adults, ages 30 and above from National Health and Nutrition Examination Survey III. Receiver Operating Characteristic curves and Cox Proportional Hazard models for the whole sample and for stratified age groups were used to compare how well Allostatic Load, Framingham Risk Score, and Biological Age predict ten-year all-cause and disease-specific mortality in the sample, for whom there were 1,076 deaths over 96,420 person years of exposure. RESULTS: Overall, Biological Age predicted 10-year mortality more accurately than other measures for the full age range, as well as for participants ages 50 to 69 and 70+. Additionally, out of the three measures, Biological Age had the strongest association with all-cause and cancer mortality, while the Framingham Risk Score had the strongest association with CVD mortality. CONCLUSIONS: Methods for quantifying biological risk provide important approaches to improving our understanding of the causes and consequences of changes in physiological function and dysregulation. Biological Age offers an alternative and, in some cases a more accurate summary approach to the traditionally used methods, such as Allostatic Load and Framingham Risk Score.
Authors: Thomas J Wang; Philimon Gona; Martin G Larson; Daniel Levy; Emelia J Benjamin; Geoffrey H Tofler; Paul F Jacques; James B Meigs; Nader Rifai; Jacob Selhub; Sander J Robins; Christopher Newton-Cheh; Ramachandran S Vasan Journal: Hypertension Date: 2007-01-22 Impact factor: 10.190
Authors: Michelle T Duong; Brianna A Bingham; Paola C Aldana; Stephanie T Chung; Anne E Sumner Journal: J Racial Ethn Health Disparities Date: 2016-06-28
Authors: Alice Cepeda; Kathryn M Nowotny; Jessica Frankeberger; Jarron M Saint Onge; Avelardo Valdez Journal: Public Health Rep Date: 2018-08-01 Impact factor: 2.792
Authors: Daniel W Belsky; Terrie E Moffitt; Alan A Cohen; David L Corcoran; Morgan E Levine; Joseph A Prinz; Jonathan Schaefer; Karen Sugden; Benjamin Williams; Richie Poulton; Avshalom Caspi Journal: Am J Epidemiol Date: 2018-06-01 Impact factor: 4.897
Authors: Daniel W Belsky; Kim M Huffman; Carl F Pieper; Idan Shalev; William E Kraus Journal: J Gerontol A Biol Sci Med Sci Date: 2017-12-12 Impact factor: 6.053
Authors: Daniel C Parker; Bryce N Bartlett; Harvey J Cohen; Gerda Fillenbaum; Janet L Huebner; Virginia Byers Kraus; Carl Pieper; Daniel W Belsky Journal: J Gerontol A Biol Sci Med Sci Date: 2020-09-16 Impact factor: 6.053