Disability and co-morbidity in relation to frailty: How much do they overlap? Olga Theou a , Michael R.H. Rockwood a , Arnold Mitnitski a,b , Kenneth Rockwood a, * a Geriatric Medicine Research Unit, Department of Medicine, Division of Geriatric Medicine, Dalhousie University and Capital District Health Authority, Halifax, Nova Scotia, Canada b Department of Mathematics and Statistics, Dalhousie University, Halifax, Nova Scotia, Canada 1. Introduction Although many older adults report being healthy, 91% have one or more chronic conditions, 40% live with a disability and about 23% are frail (Public Health Agency of Canada, 2006; Rockwood et al., 2011). Frailty is non-controversially understood as increased vulnerability to adverse health outcomes of people of the same chronological age (Rockwood et al., 1994; Theou and Kloseck, 2007; Abellan van Kan et al., 2010; Clegg and Young, 2011). The consideration of older adults’ frailty status is fundamental to their care (Theou and Rockwood, in press). For example, a severely frail 70-year-old person may not survive an aggressive medical treatment such as a major surgery, even though they are comparatively young, and may benefit more from innovative processes of care. Likewise, a fit 84-year-old might well withstand such a procedure despite being older. Frailty’s operationalization varies chiefly between groups who follow what is known as the ‘‘phenotypic’’ approach (Fried et al., 2001; Bergman et al., 2004) versus those who follow a deficit accumulation approach to defining frailty (Mitnitski et al., 2001; Kulminski et al., 2006). The frailty phenotype of five items (Fried et al., 2001) – sometimes fewer (Ensrud et al., 2009) – holds that while frailty is related to disability and co-morbidity, it is conceptually distinct (Fried et al., 2004). For this reason, some commentators recommend against including disability and co- morbidity markers as part of any frailty definition (Fried et al., 2004; Abellan van Kan et al., 2008). The deficit accumulation approach views frailty as a stochastic process in a redundant system of multiply dependent items which on average is accumulating deficits that impair physiological reserve (Rockwood et al., 2010). For this reason, in terms of understanding system behavior, knowing exactly what is wrong is less important than knowing how many things have gone wrong (Mitnitski et al., 2005). The deficit accumulation approach to frailty recommends against excluding items a priori, as long as each individual items included in a FI meets the criteria to be a deficit: item should increase with age but not become saturated (i.e. not be represent in everyone by some comparatively young age), be associated with an adverse outcome, have <5% missing data and occur at some Archives of Gerontology and Geriatrics 55 (2012) e1–e8 A R T I C L E I N F O Article history: Received 23 November 2011 Received in revised form 20 February 2012 Accepted 1 March 2012 Available online 28 March 2012 Keywords: Aging Frail Disability Co-morbidity A B S T R A C T The purpose of this study was to examine the association of disability and co-morbidity with frailty in older adults. 2305 participants aged 65+ from the second wave of the Canadian Study of Health and Aging (CSHA), a prospective population-based cohort study, comprised the study sample. Following a standard procedure, two different frailty index (FI) measures were constructed from 37 deficits by dividing the recorded deficits by the total number of measures. One version excluded disability and co-morbidity items, the other included them. Time to death was measured for up to five years. Frailty was defined using either the frailty phenotype or a cut-point applied to each FI. Of people defined as frail using the frailty phenotype, 15/416 (3.6%) experienced neither disability nor co-morbidity. Using 0.25 as the cut-point score for the FI (without disability/co-morbidity) resulted in 101/1176 (8.6%) frail participants that had neither disability nor co-morbidity. Activities of daily living (ADL) limitations and co-morbidities occurred more often among people with the highest levels of frailty. The first ADLs to become impaired with increasing frailty were bathing, managing medication, and cooking with more than 25% of older adults with a FI score (without disability/co-morbidity) >0.22 experiencing dependency on them. The hazard ratio (HR) per 0.1 increase in FI score was 1.25 (95% CI: 1.20–1.30) when disability and co- morbidity were included in the index and 1.21 (1.16–1.25) when they were not included. In conclusion, disability and co-morbidity greatly overlap with other deficits that might be used to define frailty and add to their ability to predict mortality. ß 2012 Elsevier Ireland Ltd. All rights reserved. * Corresponding author at: Centre for Health Care of the Elderly, Capital District Health Authority, 1421-5955 Veterans’ Memorial Lane, Halifax, Nova Scotia, Canada B3H 2E1. Tel.: +1 902 473 8687; fax: +1 902 473 1050. E-mail address: Kenneth.Rockwood@dal.ca (K. Rockwood). Contents lists available at SciVerse ScienceDirect Archives of Gerontology and Geriatrics jo ur n al ho mep ag e: www .elsevier .c om /lo cate/ar c hg er 0167-4943/$ – see front matter ß 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.archger.2012.03.001