A NETWORK PERSPECTIVE OF AGING AND FRAILTY. (11th November 2018)

• Record Type: Journal Article
• Authors: Farrell, S; Mitnitski, A; Rockwood, K; Rutenberg, A
• Is Part Of: Innovation in aging. Volume 2(2018)Supplement 1
• Page Start: 713
• Page End: 714
• Publication Date: 2018-11-11
• Subjects: Aging -- Periodicals; Gerontology -- Periodicals; 612.67
• DOI: 10.1093/geroni/igy023.2643
• Languages: English
• ISSNs: 2399-5300

Abstract

Abstract: We computationally model aging individuals as a network of nodes representing health attributes. Nodes stochastically damage to form health deficits. Damage rates depend on the state of neighboring nodes to represent interactions between physiological systems. The Frailty Index (FI) is the proportion of health deficits in a set of health attributes, or the proportion of damaged nodes in a subset of the network. Our model demonstrates known patterns of human aging, such as Gompertz's law of mortality and the average trajectory of the FI, without any assumption of programmed aging. We use our model to better understand factors that influence the health trajectories of individuals. Different types of health attributes can be represented in the model as different classes of network nodes. This allows us to model FI-clinical, a FI using self-reported or clinically diagnosable health attributes, and FI-lab, a FI using biomarkers or laboratory tests. Damage to nodes within these FIs affect mortality and the accumulation of deficits in different ways. Many differences in the behavior of the FI-clinical and FI-lab from the CSHA and NHANES studies can be explained in terms of the structure of the network underlying health attributes. More generally, our computational model allows us to understand how the damage occurring throughout an individual's life influences their health and lifespan.