Considering the Measurement of Frailty

What is frailty, the state of being physically frail? As is the case for the related - but not identical - state of being old we know it if we see it, but a single rigorous and commonly agreed upon measurable definition of frailty remains elusive. Not for lack of trying, of course; there are numerous published and proposed scales of measurement, and debate over their utility. The paper referenced here argues for a better language and categorization of frailty in aging, pointing out that that the present clinical applications of scientific knowledge are not what they might be, while new approaches are at this point just as likely to muddy the waters as clear things up.

The treatment of aging and its many degenerative conditions by the medical profession has a foot in two worlds. There is the established world of observation, simple measures of function used for decades, the judgement calls of medical professionals based on sparse data, the acceptance that there is little that can be done but assess the downward spiral. Within this world there is still as much reluctance to measure aging in common practice, as there is a movement towards better use of existing tools to pin down aging and be more rigorous. That old way of doing things now coexists with increasingly sophisticated by quite narrow applications of modern biotechnology that accurately measure aspects of, and to some degree treat, specific age-related dysfunctions. Despite representing sizable steps forward these new technologies have yet to give rise to a biomarker for biological age - a rigorous measure of damage and dsyfunction - despite some promising ongoing lines of research such as analysis of DNA methylation patterns. Such a biomarker would produce unambiguous results based on the state of cellular biochemistry rather than the present commonly used measures such as grip strength, recollection, or walking ability.

The practice of medicine is constantly in transition in this age of rapid progress in biotechnology. Papers such as the work referenced here are one manifestation of the tensions and necessary organization involved in the departure of old, established practice taking place concurrently with the advent of unfinished works-in-progress that will one day become the future standard of care.

Frailty: a tale of two concepts

Frailty in older adults is most often defined as a late-life vulnerability to adverse health outcomes. Although frailty consensus work has focused on physical frailty, a theoretically based construct built around a core group of activity-based and strength-based measurements, the frailty index (FI), has emerged as a useful strategy to measure risk for mortality and other adverse health outcomes in older adults. The FI utilizes simple counts of up to 71 co-morbidities, laboratory measures, and social and functional declines (termed deficits) to construct a score. Proponents of this approach have noted that the component measures are interchangeable, the approach can be applied in bedbound or ambulatory populations, and fewer variables can successfully predict mortality than originally proposed.

In the present study, the authors develop an aging-related biological index that utilizes 40 biological measures found to be age-associated in the Newcastle 85+ study. This index contains measures ranging from telomere length to induced cytokine production from isolated lymphocytes to the components of a complete blood count. Although prior studies of older adults have developed indices focused on clinical laboratory measurements, the authors are to be commended for working toward the development of an index that attempts to assess biological age and associated risk through biological measurements. Their findings regarding the complementarity of clinical and biological measures for predicting mortality risk are potentially important: One can imagine, for example, the identification of older adults who "look" healthy but may benefit from interventions to address significant latent risk factors.

The approach is pragmatic in that measures can be for the most part interchanged with other measures without substantial change in the predictive ability of the tools. There are likely a large number of age-related aggregated biological precursors that drive frailty and late-life decline, and the systemic effects as measured by risk to adverse outcomes can in part be detected through an index approach. Hence the approach can be used to predict outcomes, and the combination of FI and a biomarker-based frailty index (FI-B) seems to hold considerable promise to this end. The approach also likely allows the tracking of vulnerability in a reasonable way.

Despite this flexibility in measurement and substantial predictive ability of a high index score, notable obstacles remain if this approach is to be further developed as a true biological aging index or measure of systemic effects. For example, it is not clear that systemic effects are being measured unless index components are selected and validated vis a vis a system. Moreover, including a preponderance of age-associated clinical measures may result in an assessment of chronic disease states rather than aging per se. The FI-B construction, and the authors' characterization of frailty as "a state of increased risk, compared with others of the same age," suggests that any marker conferring risk of mortality (or possibly other adverse geriatric outcomes) contributes usefully to frailty measurement. We, and others who consider frailty as a specific physiological state with a definable phenotypic presentation, would disagree.

The specificity embedded within the phenotype approach offers benefits if the goal is to elucidate mechanisms and physiological etiology. To approach such a goal scientifically, theories describing plausible processes by which frailty arises, and how they are linked to one another, are needed. Indices (psychometrically: as opposed to "scales") fail to provide this. Moreover, the more heterogeneously a large collection of variables arises, the higher the risk of masking a key driver that can be targeted by specific interventions. In sum, various conceptualizations of frailty have complementary strengths; however, if etiology is to be elucidated and targeted interventions are to be pursued, we believe the field would benefit from more strongly distinguishing disparate concepts which currently share the single label of "frailty."

Comment Submission

Post a comment; thoughtful, considered opinions are valued. New comments can be edited for a few minutes following submission. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.