A comparison of frailty indexes for prediction of adverse health outcomes in an elderly cohort
Introduction
Frailty is a condition that makes elderly persons highly vulnerable to adverse health outcomes (Hogan et al., 2003). Identification of frail persons is of interest for both researchers and clinicians, but a uniformly accepted operational definition for frailty is still lacking (Martin and Brighton, 2008).
According to the physical phenotype of frailty that Fried et al. (2001) developed from the Cardiovascular Health Study (CHS) cohort, frailty is defined as a cluster of five measurements related to muscular mass and strength, taken as indicators of physical fitness. As a construct that neatly distinguishes frailty from disability and comorbidity, the physical phenotype is designed to capture pre-clinical states of pure physiologic vulnerability and has provided very valuable insights about frailty's biology and physiology (Walston et al., 2006). A simplified, 3-item version of the CHS index has been developed from the SOF cohorts and proposed for clinical use (Ensrud et al., 2008, Ensrud et al., 2009).
Geriatricians, however, are most often faced with full-blown frailty, when physical vulnerability has already progressed toward functional impairment and its clinical repercussions are heavily affected by, and often inseparably interlaced with, comorbidity, sensory loss, and psychosocial status (Walston et al., 2006, Whitson et al., 2007, Hubbard et al., 2009). This stage of frailty may be best described as an accumulation of multifactorial deficits (Hogan et al., 2003, Walston et al., 2006, Martin and Brighton, 2008). Several cumulative indexes exist but, due to the high number of included items, their lack of standard norm, or the need of specific equipment for their measurement, most of them are challenging for clinical use (Bergman et al., 2007).
Using data from the CSBA, an Italian population-based cohort of elderly people, we previously proposed a cumulative frailty index including only nine standardized measurements (Ravaglia et al., 2008). In this paper, data from the same cohort were used to compare the ability of the CSBA and mSOF index for prediction of several adverse outcomes known to be related to frailty.
Section snippets
Subjects and setting
The CSBA is a population-based prospective survey of Italian elderly people aimed to provide epidemiological data about age-related cognitive disorders, as detailed elsewhere (Ravaglia et al., 2005). Briefly, in 1999/2000, 1016 (75%) of the 1353 individuals aged 65 years and older residing in the Conselice municipality (Emilia Romagna Region, Northern Italy) were enrolled for baseline examination. In 2003/2004, survivors underwent a follow-up examination. Informed consent for collection and use
Results
The characteristics of the baseline cohort are summarized in Table 1. Frailty as defined by CSBA index was present in more than one third of participants whereas, according to the mSOF index, only one tenth of the cohort was frail. Subjects identified as frail according to the CSBA index were older (79.6 ± 7.1 vs. 71.7 ± 5.1 years, p < 0.001), more frequently men (52.0% vs. 40.1%, p < 0.001), and less educated (41.9% with ≤3 years of education vs. 30.7%, p < 0.001) than their non-frail counterparts.
Discussion
This study of an elderly population-based cohort supports the view that full-blow frailty is more common than pure physical vulnerability (Bergman et al., 2007). According to previous investigations, prevalence of frailty ranges from 7.2% to 25.7% when using the physical phenotype (Fried et al., 2001, Gill et al., 2006), but rises from 16% to 43% when using cumulative frailty indexes (Puts et al., 2005, Rockwood et al., 2005). Although comparisons are hampered by huge differences in
Conclusion
The data show that different approaches to frailty result in different estimates for the prevalence of individuals at risk and different predictive accuracy for the outcomes of interest. The data also support the hypothesis that a cumulative index based upon the principle of multidimensional geriatric assessment may provide a better estimation of frailty-related adverse health outcomes than a unidimensional index exclusively focused on muscular fitness. Replication and extension of our findings
Conflict of interest statement
None.
Acknowledgement
This work was supported by the Basic Oriented Research grant from the University of Bologna.
References (26)
- et al.
Development and validation of a geriatric depression screening scale: a preliminary report
J. Psychiatr. Res.
(1983) - et al.
Frailty: an emerging research and clinical paradigm. Issues and controversies
J. Gerontol. A: Biol. Sci. Med. Sci.
(2007) - et al.
An evaluation of screening measures for cognitive impairment after stroke
Age Ageing
(2002) - et al.
Anthropometric approaches to the nutritional assessment of the elderly
- et al.
Comparison of 2 frailty indexes for prediction of falls, disability, fractures, and death in older women
Arch. Intern. Med.
(2008) - et al.
A comparison of frailty indexes for prediction of falls, disability, fractures, and mortality in older men
J. Am. Geriatr. Soc.
(2009) - et al.
Frailty in older adults: evidence for a phenotype
J. Gerontol. A: Biol. Sci. Med. Sci.
(2001) - et al.
Transitions between frailty states among community-living older persons
Arch. Intern. Med.
(2006) - et al.
Models, definitions, and criteria of frailty
Aging Clin. Exp. Res.
(2003) - et al.
Characterising frailty in the clinical setting—a comparison of different approaches
Age Ageing
(2009)