Defining Frailty in a way that works.
There is no single criterion standard for the measurement of frailty. Frailty has been defined as both a state of exaggerated vulnerability and a multidimensional syndrome. There are many different measures for the construct of frailty. These may be used alone or in combination in research settings, depending on the objectives of the researchers. The following is a brief overview of different categories of frailty measures with representative examples, followed by a table that compares their characteristics.
Physical Frailty: In research, no measure has been more widely used to capture the syndrome of frailty than the Frailty Phenotype (FP) [1], which is based on five physical traits – slow gait speed, weakness by grip strength, subjective exhaustion, unintended weight loss, and low physical activity. The FP has not been well-adopted in clinical settings.
Biological Markers: A physiologic basis for frailty has long been hypothesized and tested in relation to measures such as the PF. However, “no marker has been proven to be individually of sufficient diagnostic and prognostic capacity to be valid in clinical settings”, as summarized in an excellent recent review [2].
Accumulation of Deficits: The state of frailty can be constructed in mathematical terms using this time-honored model. The Frailty Index [3] is very attractive in research settings because it is flexible to the available dataset. The FI is calculated as the quotient of deficits present from a pre-determined list of at least 30 candidate items. It has also been operationalized in clinical terms and implemented successfully as the electronic Frailty Index [4].
Judgment-based: When clinicians have the confidence and experience to make judgments about the presence and severity of frailty, a judgment-based measure such as the Clinical Frailty Scale [5], allows them to rapidly record that impression, based on a set of nine grades of frailty which have descriptive anchors.
Self-Report: Although few older adults living with frailty would characterize themselves this way, their response to questions on related topics may be an important starting point in case-finding. Self-report measures such as the PRISMA 7 [6], Tilburg Frailty Indicator [7], and FRAIL [8] are good examples of how such a self-report measure could be completed before an encounter between clinician and patient.
Performance-based: Case-finding can be also accomplished by observing the performance by an individual. For example, grip strength and gait speed, both items in the FP, have been used alone for this purpose. Gait speed in particular has been shown to be a powerful predictor of adverse outcomes associate with frailty [9].
Multidimensional Measures: In contrast to measures of physical frailty, these measures are intended to capture the multidimensional and dynamic aspect of frailty. Whereas the Frailty Index based on Comprehensive Geriatric Assessment (FI-CGA) requires prior CGA to be completed, the EFS and Groningen Frailty Indicator can be administered directly before such an assessment. These measures may comprise a combination of self-report, performance-based, and clinician rated items.
Multidimensional measures offer frailty case-finding, estimation of severity, and component definition.
A Description of Different Categories of Frailty Measures with Examples
Characteristics of Frailty Measures by Category
The following table may assist researchers in determining which frailty measures, alone or in combination are best suited to their study.
What about standardization & future implementation?
When implementing a frailty measure in a health care setting, the following are important considerations. These are especially important when the intended use of the frailty measure is systematic rather than opportunistic or ad-hoc. For a more thorough discussion, please see a recent review of what it would take to implement frailty measures in the Canadian Health Care System [13]. The Frailty Outcomes ConsensUS (FOCUS) Project is seeking participation in a Delphi-based process to work towards common data elements and outcome measures.
For more information about FOCUS: www.cfn-nce.ca/news/the-cfn-led-frailty-outcomes-consensus-focus-project-is-looking-for-your-input/
Standardization of Measurement: Each item should be clearly defined so that there is consistency in measurement and interpretation.
Specifications of data standards and coding rules: Coding rules and data standards are necessary for the measure to be used in electronic records or by software vendors, especially if larger scale data analysis is anticipated.
Training: A training strategy that matches the measure is crucial for spread of the instrument widespread ongoing use.
Reporting Standards: Necessary if the data are to be aggregated for use at the organizational, regional. Provincial or national level.
Cross-sector consistency: Older adults living with frailty frequently transition between different healthcare sectors. Research that aims to inform cross-sector comparisons should employ appropriate measures.
Data sharing: If there is a cross-sectoral approach, then data from frailty measures will need to be easy to access by parties in relevant sectors.
Timing for frailty as an outcome: If frailty measures are used as an outcome in research, then there should be standardization in the timing of assessments, to allow comparisons between studies.
Data quality: Both researchers and clinicians should consider continuous, real-time mechanisms to ensure that frailty data continues to measure what it is intended to measure in a reliable way.
References
Fried, L.P., et al., Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci, 2001. 56(3): p. M146-56.
Vina, J., et al., Biology of frailty: Modulation of ageing genes and its importance to prevent age-associated loss of function. Mol Aspects Med, 2016. 50: p. 88-108.
Mitnitski, A.B., A.J. Mogilner, and K. Rockwood, Accumulation of deficits as a proxy measure of aging. ScientificWorldJournal, 2001. 1: p. 323-36.
Clegg, A., et al., Development and validation of an electronic frailty index using routine primary care electronic health record data. Age Ageing, 2016. 45(3): p. 353-60.
Rockwood, K., et al., A global clinical measure of fitness and frailty in elderly people. Cmaj, 2005. 173(5): p. 489-95.
Raiche, M., R. Hebert, and M.F. Dubois, PRISMA-7: a case-finding tool to identify older adults with moderate to severe disabilities. Arch Gerontol Geriatr, 2008. 47(1): p. 9-18.
Gobbens, R.J., et al., The predictive validity of the Tilburg Frailty Indicator: disability, health care utilization, and quality of life in a population at risk. Gerontologist, 2012. 52(5): p. 619-31.
Morley, J.E., T.K. Malmstrom, and D.K. Miller, A simple frailty questionnaire (FRAIL) predicts outcomes in middle aged African Americans. J Nutr Health Aging, 2012. 16(7): p. 601-8.
Pamoukdjian, F., et al., Measurement of gait speed in older adults to identify complications associated with frailty: A systematic review. J Geriatr Oncol, 2015. 6(6): p. 484-96.
Rolfson, D.B., et al., Validity and reliability of the Edmonton Frail Scale. Age Ageing, 2006. 35(5): p. 526-9.
Jones, D.M., X. Song, and K. Rockwood, Operationalizing a frailty index from a standardized comprehensive geriatric assessment. J Am Geriatr Soc, 2004. 52(11): p. 1929-33.
Steverink, N.S., Joris & Schuurmans, Hanneke & Van Lis, M., Measuring frailty: Developing and testing the GFI (Groningen Frailty Indicator). Gerontologist, 2001. 41: p. 236-237.
Rolfson, D.B., et al., Implementing Frailty Measures in the Canadian Healthcare System. J Frailty Aging, 2018. 7(4): p. 208-216.
