There is no shortage of theorizing on the nature of aging: its biochemical causes; its evolutionary origins; how it progresses; how to measure it. In any era in which thinking is cheap and life science research is expensive, there will be a lot more theorizing than data. While the tools of biotechnology cost less than ever, and the price continues to fall even as capabilities increase radically, I think it arguably the case that we are still in the era of relatively cheap thought and relatively expensive research.
One area in which theory and modeling has over the years found its way to practical use in clinical medicine is in the construction of measures of aging based on a straightforward combination of measures, such as grip strength, markers of inflammation, and so forth. Geriatric medicine has and continues to make widespread use of these assessments of frailty. A great deal of work on measures of aging still takes place, as illustrated by the growth of epigenetic clocks on the one hand and more complex algorithmic combinations of simple health metrics on the other. The work here is an example of the latter, with the choice of metrics and their combination driven by a systems biology view of aging.
Even to the untrained eye it has always been apparent that different people age differently. Subjective evaluation of age rather accurately assesses the ravages of time and coincides quite adequately to more objective metrics. Nevertheless, we would like to be able to reference such objective measures to examine in greater detail the dimensions of aging. The dimensions of aging encompass at least three different aspects. The first incorporates prediction of survival or mortality. In other words, we want to be able to relate a process, aging, to an outcome, longevity. This has long been a domain of aging research, and it continues to engage biodemographers. The second attempts to relate an aging process to the ability to function. So-called healthy aging derives from this approach. Finally, the need to evaluate potential therapies or interventions to extend this healthspan is yet another dimension.
Deficit indices, also known as frailty indices, constitute an uncomplicated way in which to describe the behavior of a complex aging system. Deficit indices have a long history in human aging research and in geriatrics. A deficit index is constructed from a number of signs, symptoms, marks, and manifestations. The number can be relatively small, about twenty, or much larger, as long as it is statistically sufficient. These deficits should encompass many different body or physiological systems. The deficit index arises by summing the deficits counted and dividing by the total number of deficits assessed. Increasing the number of deficits scored improves deficit index performance.
Recently, the deficit index has acquired a strong theoretical underpinning. The deficits are represented by the components of a network, in which they can be damaged or undamaged (deficits per se). By definition, the components are connected by edges. Some of them have more edges than others, performing a more critical role in the network. Damage in this network, whether partial or complete, is propagated across the network or system because of the edges. This rational, systems biology-based nature of the deficit index distinguishes it from other quantitative measures of biological age. In addition, the deficit index is uncomplicated mathematically, as opposed to most of the other measures, and it predicts mortality without the incorporation of chronological age as one of its items.
We have constructed a deficit index we call frailty index-34 (FI34), consisting of 34 health and function variables. The reference to frailty in the name stresses the relevance of the index as a measure of relative health. FI34 is a good predictor of mortality, so it is a measure of biological age. It increases exponentially with calendar age, as we would expect of a predictor of mortality. Moreover, it distinguishes different patterns of aging, and it is heritable. FI34 also captures the individual variability or heterogeneity of aging among individuals. Although constantly increasing with chronological age across a population, it shows variation among individuals in cross-section and longitudinally.