Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Oct 1.
Published in final edited form as: West J Nurs Res. 2013 May 15;35(9):1128–1152. doi: 10.1177/0193945913487608

Correlates of Frailty Among Homeless Adults

Benissa E Salem 1, Adeline M Nyamathi 1, Mary-Lynn Brecht 1, Linda R Phillips 1, Janet C Mentes 1, Catherine Sarkisian 1, Barbara Leake 1
PMCID: PMC3759620  NIHMSID: NIHMS486047  PMID: 23676627

Abstract

Frailty, a relatively unexplored concept among vulnerable populations, may be a significant issue for homeless adults. This cross-sectional study assessed correlates of frailty among middle age and older homeless adults (N = 150, 40–73). A Pearson (r) bivariate correlation revealed a weak relationship between frailty and being female (r = .230, p < .01). Significant moderate negative correlations were found between frailty and resilience (r = −.395, p < .01), social support (r = −.377, p < .01), and nutrition (r = −.652, p < .01). Furthermore, Spearman’s rho (rs) bivariate correlations revealed a moderate positive relationship between frailty and health care utilization (rs = .444, p < .01). A stepwise backward linear regression analysis was conducted and in the final model, age, gender, health care utilization, nutrition, and resilience were significantly related to frailty. Over the next two decades, there is an anticipated increase in the number of homeless adults which will necessitate a greater understanding of the needs of this hard-to-reach population.

Keywords: homeless, frailty, vulnerability, health disparities

Homeless populations are aging in the United States (Hahn, Kushel, Bangsberg, Riley, & Moss, 2006) and evidence suggests will double by 2050 (Sermons & Henry, 2010). In San Francisco, 33% of the homeless population were above 50 years of age (Hahn et al., 2006) and in Los Angeles, nearly one quarter of the 51,000 homeless adults were 55 to 61 years of age (Los Angeles Housing Services Administration [LAHSA], 2011).

Population aging may be associated with clinical geriatric disorders, namely, frailty, wherein an accumulation of deficits (Rockwood & Mitnitski, 2007, 2011) in physical, psychological, and social domains may lead to adverse outcomes such as disability and mortality (Gobbens, van Assen, Luijkx, & Schols, 2012). Among homeless persons in particular, adverse life events such as trauma (Hamilton, Poza, & Washington, 2011), drug and alcohol abuse (Fountain, Howes, Marsden, Taylor, & Strang, 2003), incarceration (Greenberg & Rosenheck, 2008), along with physical and mental health conditions (Brown, Kiely, Bharel, & Mitchell, 2012; Fazel, Khosla, Doll, & Geddes, 2008; A. Nyamathi et al., 2011), may likewise place this population at greater risk for frailty, and adverse outcomes such as hospitalizations, incident falls, and premature mortality (Fried, Ferrucci, Darer, Williamson, & Anderson, 2004; Puts, Lips, Ribbe, & Deeg, 2005).

In a homeless community that has a high burden of disease and disabilities, identifying frailty among an already vulnerable population may help us to understand these interrelationships. Thus, the purpose of this study is first to identify correlates of frailty among homeless adults and discuss these initial findings and implications.

The concept of frailty has been aptly debated in the literature for decades; more recently, it has been defined as a decline in multiple physiological systems, a geriatric syndrome (Gielen et al., 2012), or a nonspecific state of vulnerability (Fulop et al., 2010). There is extensive literature to operationalize frailty measures based on conceptual differences (Fried et al., 2001; Searle, Mitnitski, Gahbauer, Gill, & Rockwood, 2008). Two popular schools of thought predominate the literature; one identifies frailty as a clinical syndrome with specific hallmark characteristics (Fried et al., 2001), whereas others define frailty as an accumulation of deficits that include symptoms, signs, and disease classifications, often leading to adverse outcomes (Mitnitski, Mogilner, MacKnight, & Rockwood, 2002). Based on the Fried phenotype, prefrailty is presence of two out of five frailty signs (shrinking, weakness, exhaustion, slowness, low activity, whereas frailty is the presence of at least three or more of those signs; Fried et al., 2001).

Frailty prevalence rates differ based on the populations tested and the operational definition utilized. In a study of relative fitness of individuals in Canada across the life span (N = 14,713, 15–102 years of age), findings revealed that 7.2% of the population was frail at baseline and the frail segment was more likely to be hospitalized when compared with their nonfrail counterparts (Rockwood, Song, & Mitnitski, 2011). In another population-based study using a modified frailty index (FI) among older Mexican Americans, 37.1% of the population was frail, 33.3% of the population was pre-frail and 29.6% of the population was not frail (Aranda, Ray, Snih, Ottenbacher, & Markides, 2011). Another study found that the prevalence of frailty was nearly 7% among community dwelling older adults (Fried et al., 2001), and when applying the same criteria to a Boston-based homeless sample, aged 50 to 69 years, the frailty prevalence was 16% (Brown et al., 2012).

Understanding frailty among homeless population necessitates clarification regarding frameworks as previous models have described clinical pathways of frailty such as underlying alterations, clinical features, and adverse outcomes (Fried & Walston, 2003). Furthermore, some focus on age-related physiologic changes, which include sarcopenia, neuroendocrine dysregulation, and immune dysfunction (Fried & Walston, 2003). Some aspects that are lacking in the literature include a guiding framework for homeless and otherwise vulnerable populations.

Frailty Framework Among Vulnerable Populations (FFVP)

The FFVP is a modification of the Integral Conceptual Model of Frailty (ICMF; Gobbens et al., 2012), the Working Frailty Framework model (Bergman et al., 2004), the vulnerable populations model (Flaskerud & Winslow, 1998), and biological models of frailty (Fried & Walston, 2003). The FFHVP is the theoretical framework that has evolved from empirical research and consultation from frailty experts (R. Gobbens, personal communication, September 23, 2012) in an effort to characterize situational, health-related, behavioral, resource, biological, and environmental factors which contribute to physical, psychological, and social frailty domains and ultimately contribute to adverse outcomes such as disability, hospitalization, health care dependency, and death.

In this model, situational variables included race/ethnicity, gender, income, education, marital status, physical, sexual, verbal victimization, and homelessness. Behavioral factors such as drug and alcohol use may likewise be significant issues among this population (A. Nyamathi, Hudson, Greengold, & Leake, 2012; A. M. Nyamathi, Leake, & Gelberg, 2000) affecting health-related factors and increasing health care utilization (Hahn et al., 2006; Stein, Andersen, Robertson, & Gelberg, 2012). Resource factors include resilience which may influence the aforementioned antecedents. The domain of frailty is composed of physical, psychological, and social domains, which may impact each other. Physical frailty may encompass slow walking, decreased grip strength, and an overall decline in physical functioning, whereas psychological frailty may be composed of a decline in cognition, and coping, all of which may affect physical and social frailty. Alternatively, social frailty may affect physical and psychological frailty. Adverse outcomes of frailty included disability, hospitalization, health care dependency, and premature mortality.

Investigators have noted that homeless populations have a substantial disease burden (Garibaldi, Conde-Martel, & O’Toole, 2005). Health conditions related to aging, namely, chronic health conditions (Garibaldi et al., 2005), depression, and disabilities are all significant issues (LAHSA, 2011). Furthermore, visual and hearing impairment, functional limitations, and cognitive impairments (Brown et al., 2012) may likewise place older adults at greater vulnerability.

Research suggests that food insecurity is a formidable challenge among homeless populations, compromising nutrient intake due to lack of sufficient food (Dachner & Tarasuk, 2002). Baggett et al. (2011) studied food insufficiency and health service utilization among a national sample of homeless adults and found that 25% did not get enough to eat. In addition, among those who were chronically homeless, prevalence of food insufficiency was 45.5% and 37.5% among those who had been physically or sexually abused in the past year (Baggett et al., 2011). Food insufficiency was similarly linked to increased health care utilization; in particular, those who were food insufficient were more likely to be hospitalized in the last year when compared with those who were not food insufficient (46.3% vs. 30.3%; Baggett et al., 2011).

Poor social support has been linked to increasing frailty (Woo, Goggins, Sham, & Ho, 2005). Research indicates that homeless populations can similarly be socially isolated (Hwang et al., 2009). One study focused on multidimensional social support among homeless adults (N = 544) in Canada and differences were noted between perceptions of support versus actually receiving support. In particular, 62% perceived access to financial support, 51% perceived access to instrumental support, and 60% perceived access to emotional support; however, only 7% were accompanied to a health care provider by a family or friend (Hwang et al., 2009). Understanding the influence of social support and frailty among homeless persons will enable a greater understanding of the dimensional and intersecting relationships.

Frail adults and homeless adults utilize a large majority of health care services (Hahn et al., 2006). Hahn et al. (2006) studied aging trends more than 14 years among homeless adults (N = 3,534) and found that median age increased, along with emergency department visits (p = .01), staying overnight in a hospital (p < .001), and mental health hospitalization (p < .001). Authors assert that due to aging, chronic health conditions will be predominant among homeless adults (Hahn et al., 2006). Furthermore, women have been found to be frailer when compared with men (Garre-Olmo, Calvo-Perxas, Lopez-Pousa, de Gracia Blanco, & Vilalta-Franch, 2013; Goggins, Woo, Sham, & Ho, 2005; Syddall et al., 2010). While chronological age may serve as a proxy of a person’s vulnerability to frailty (Bergman et al., 2007), in effect, age often becomes a principal risk factor for disease. Seminal authors have indicated that homeless populations age prematurely, in fact, generally 10 to 20 years older than their chronological age (Gelberg, Linn, & Mayer-Oakes, 1990).

Prolonged exposure to stress may induce premature aging (Epel et al., 2004) described as weathering, which may be of significant concern for those living in poverty (Geronimus et al., 2010; Geronimus, Hicken, Keene, & Bound, 2006), homeless, and on the streets. One author compared immune markers of homeless (n = 40) and nonhomeless housed counterparts (n = 40) in Spain and found that when compared with nonhomeless counterparts, homeless adults had impaired interleukin-2 and lower plasma total antioxidant capacity (Arranz, de Vicente, Munoz, & De la Fuente, 2009). Homeless adults may have altered immune systems and increased oxidative stress (Arranz et al., 2009). One biological marker, telomere length, may serve as an indicator of cellular senescence and may be affected by consistent stress activation (Epel et al., 2004; Kotrschal, Ilmonen, & Penn, 2007). Some data suggest that telomere length is affected by chronic stress in socially and economically vulnerable adults (Geronimus et al., 2010).

On the basis of existing empirical literature, we hypothesized that there would be a positive relationship between frailty and (a) chronological age, (b) length of time homeless, (c) health care utilization, (d) illicit drug use, and (e) women would be more frail when compared with men. We also hypothesized that we found (f) an inverse relationship between resilience, (g) nutrition, and (h) social support and frailty.

Method

Design

In this cross-sectional study, homeless adults from Los Angeles were enrolled and administered a survey designed to assess correlates of frailty. The study was approved by the University Human Subjects Protection committee, with data collected from February to May 2012.

Sample

The sample included 150 homeless adults; based on the power analysis, this would allow detection of a small to medium (.22) effect size at an alpha of .05 and power of .80. Participants who were eligible for the study if they were (a) aged 40 or above, (b) free of evidence of acute psychotic hallucinations and psychosis, (c) English-speaking, and (d) considered homeless. “Homeless” was operationally defined as an individual who lacks a fixed, regular, and adequate nighttime residence, and who has a primary nighttime residence that was a supervised publicly or privately operated shelter designed to provide temporary living accommodations (U.S. Department of Housing and Urban Development, 1995). Participants were recruited from three participating homeless day center drop-in sites on skid row and one residential drug treatment (RDT) facility, which provides temporary shelter for homeless adults on parole or probation within the same perimeter.

Procedure

The principal investigator (PI) established partnerships with the research sites, and upon obtaining human subjects institutional board approval, flyers were posted in common areas at the sites. The PI made frequent announcements in the research sites during the recruitment period. In total, 313 homeless adults showed interest and approached the PI in a private and confidential area of each respective agency. If interest continued, a brief screening questionnaire was administered by the PI to assess birth year, homeless status, and sleeping arrangements in the previous night. If determined eligible, the PI set an appointment with the potential participant and subsequently completed informed consent in a quiet screened area of the facility. The PI administered the questionnaires and other assessments; sessions lasted approximately 1 hr and 30 min. At the end of the session, each participant was compensated with a US$25 gift card that could be utilized at a neighboring food vendor.

Instruments

The instrumentation was composed of measures chosen based on the FFHVP and selected empirical data linking situational, behavioral, health-related, and resource factors.

Situational Factors

Sociodemographic data were obtained related to chronological age, gender, birthplace, race/ethnicity, educational history, marital status, education, current monthly income, number of times homeless in their lifetime, living arrangements, and length of time homeless.

Health-Related Factors

Comorbid conditions were assessed using the self-reported comorbidity index (SCQ) for medical conditions, treatment, and physical limitations (Sangha, Stucki, Liang, Fossel, & Katz, 2003). The Total Problem subscale consisted of 13 questions and two open-ended spaces. Responses were coded as “yes/no” along with the presence of a condition. Higher numbers meant higher comorbid scores. The alpha coefficient for the total problem score was .91 in this sample.

Depressive symptomology was assessed using the Center for Epidemiologic Scale (CES-D; McDowell, 2006; Radloff, 1977). The scale consisted of 20 items with a 4-point response scale; responses ranged from “rarely or none of the time” to “most of the time.” Items 4, 8, 14, and 16 were reverse scored. Scores of ≥16 indicates a need for psychiatric evaluation for depressive symptoms (Weissman, Sholomskas, Pottenger, Prusoff, & Locke, 1977). The alpha coefficient for the CES-D in this sample was .89.

Physical functioning was assessed using the Medical Outcomes Study (MOS) Physical Functioning Measure (McDowell, 2006). This 10-item self-report instrument was used to determine functioning and scores ranged from 1 “limited a lot” to 3 “not limited at all.” Individuals with higher scores have better functioning. The alpha coefficient for part 1 (10 items) of physical functioning was .94.

Falls was assessed by three self-report questions about having fallen in the last year, in the last 30 days and having a fear of falls. Responses included “yes/no.” The alpha coefficient for falls was .71.

Behavioral Factors

Drug use and dependency was assessed using the Texas Christian University (TCU) Drug Screen II (Knight, Simpson, & Hiller, 2002). The 15-item self-report screening test provides an understanding of a history of heavy drug use or dependency within the last 12 months. Responses included “yes/no” to each drug mentioned. The total score ranges from 0 to 9; higher scores (≥3 or greater) indicate relatively severe drug-related problems and corresponds approximately to Diagnostic and Statistical Manual of Mental Disorders (DSM) drug dependence diagnosis. The alpha coefficient for drug use and dependency was .95.

Nutrition was assessed using the mini nutritional assessment (MNA; DiMaria-Ghalili & Guenter, 2008; Vellas et al., 1999). The screening includes questions related to weight loss in the last 3 months, mobility, psychological distress, neuropsychological problems, and body mass index (BMI). If a client scores 11 points or less in the screening (6 items; Part I), they then need to proceed to the nutritional assessment (12 items; Part II), which includes questions related to types of protein intake, use of prescription pills, pressure sores, and mid arm and calf circumference. A total score is derived that indicates whether the participant is malnourished (<17), at risk of malnutrition (17–23.5) or has a normal nutritional status (24–30). The alpha coefficient for nutrition was .70.

BMI was evaluated by measuring height and weight. Weight was assessed in pounds, whereas height was assessed in inches.

Resource Factors

Resilience was assessed using the resilience scale (Wagnild, 2009; Wagnild & Young, 1993). The 25-item index evaluates a purposeful life, perseverance, equanimity, self-reliance, and existential aloneness activities on a 7-point scale ranging from (1) “strong disagree” to (7) “strongly agree”; the total score ranges from 25–175 (Wagnild, 2009; Wagnild & Young, 1993). A higher score meant greater resilience. The alpha coefficient for resilience was .94.

Social support was assessed using the MOS Social Support Survey (MOS-SSS; Sherbourne & Stewart, 1991). Items in the four subscales were evaluated or a 5-point scale ranging from (1) “none of the time” to (5) “all of the time.” The data were transformed to a 0- to 100-point scale. Alpha coefficient for the subscales included emotional support (eight items, α = .95), tangible support (four items, α = .93), positive social interaction (four items, α = .94), and affectionate support (three items, α = .91). The overall alpha coefficient for social support was .97.

Adverse Outcomes

Disability was assessed using the Katz activities of daily living (ADL) scale (Katz & Akpom, 1976). The six-item scale takes into account bathing, dressing, toileting, transfer, continence, and feeding. The instrument uses “yes/no” scale with 1 point given for each ADL in which the participant is independent and 0 points given for each ADL in which the participant is dependent. Higher scores mean greater level of independence and lower scores signify higher dependence.

Health care utilization was assessed by self-report questions on health care utilization within the last year (e.g., seeing a health care provider, frequency of emergency department use). Respondents answered “yes/no” and number of times they sought health care services (0–35 times); a total score was added for items. A higher score indicated using more health care.

Frailty was assessed using the FI (Rockwood et al., 2011), which measures symptoms, signs, and disease classifications. The PI consulted with primary developmental investigators related to the FI items (A. Mitnitski, personal communication, November 3, 2011; K. Rockwood, personal communication, July 30, 2010), coding, and analysis (A. Mitnitski, personal communication, June 26, 2011; July 13, 2011; March 2, 2012). This study utilized an established cut point assignment as relatively fit (FI ≤ 0.03, that is, no or only one deficit), less fit (0.03 < FI ≤ 0.10), least fit (0.10 < FI ≤ 0.21), frail (0.21 < FI < .44), and most frail (FI ≥ 0.45; Rockwood et al., 2011). The alpha coefficient for the 42-item FI was .88.

Analysis

Frequencies and percents were used to describe the sample characteristics of age, gender, educational level, socioeconomic status, marital status, and length of time homeless. Variables that were not normally distributed were log-transformed and they included health care utilization, drug use, and length of time homeless. To evaluate bivariate associations between frailty and possible predictors, Pearson correlations assessed normally distributed variables, whereas Spearman rho correlations assessed independent-level variables, which were not normally distributed. Stepwise backward linear regression was selected as the method to identify independent correlates of frailty. The resulting model was confirmed using a stepwise forward regression.

Several variables were originally selected because they are significant among homeless populations (i.e., comorbid conditions, physical functioning, depressive symptomology, falls, BMI, and ADLs). However, to avoid tautological error and overlap with the dependent variable, these variables were excluded as independent variables from bivariate and multivariate analysis. Assumptions of linearity and normal distribution were checked and met. In particular, the histogram illustrated a normal distribution for the dependent variable. Likewise, scatterplots depicted linear relationship between frailty and each continuous independent variable and homoscedasticity with the points being randomly and evenly dispersed throughout the plot. Multicollinearity diagnostics were checked by determining whether the variance inflation factor (VIF) and tolerance (1/VIF) have strong linear relationships; based on collinearity diagnostics, values were not problematic and within range.

Results

Sociodemographic and Behavioral Characteristics

Table 1 presents demographic characteristics for the sample. The mean age was 52.4 (40–73; SD 6.80). Slightly more than half of the participants (59.3%) were above the age of 50 and gender was equally distributed. More than half of the sample were frail (54%) and 10% were considered most frail. With respect to race/ethnicity, the majority of participants were African American (63.3%), followed by Anglo/White/Caucasian (12%), and Hispanic/Latino (10.7%). The majority of participants was unmarried (48%) or divorced (34.7%) and most completed Grades 9 to 12 (53.3%). About one third of the sample completed some college (32%).

Table 1.

Sociodemographic Characteristics (N = 150 Homeless Adults).

Characteristic n %
Age
 ≤50 61 40.7
 >50 89 59.3
Relatively fit 4 2.6
Less fit 17 11.3
Least fit 33 22.0
Frail 81 53.3
Most frail 15 10.0
Gender
 Male 75 50.0
 Female 75 50.0
Children
 Yes 97 64.7
 No 53 35.3
Birthplace
 United States 143 96.0
 Mexico 1 0.7
 Other 6 3.3
Race/ethnicity
 African American 95 63.3
 Anglo/White/Caucasian 18 12.0
 Hispanic/Latino 16 10.76
 Mixed 11 7.3
 African American/American Indian 5 3.3
 Other 2 1.3
 American Indian 1 0.7
Marital status
 Never married or unmarried 72 48.0
 Divorced 52 34.7
 Separated 10 6.7
 Widowed 10 6.7
 Legally married 4 2.6
 Living with partner 2 1.3
Highest level of education
 ≤8th grade 11 7.3
 9th grade–12th grade/GED 80 53.3
 Some college 48 32.0
 College completion 8 5.3
 Graduate degree and professional school 3 2.0
Types of financial support received (last year)
 Pension 1 0.7
 Savings 1 0.7
 Job 2 1.3
 Casual work 6 4.0
 Unemployment check 5 3.3
 Social security insurance/social security disability 53 35.3
 Money or financial support from support network 13 8.7
 General relief (welfare) 62 41.3
 Food stamps 1 0.7
 None 6 4.0
Current monthly income
 US$0–US$100 47 31.3
 US$100–US$500 55 36.7
 US$500–US$2,000 48 32.0
Lifetime years homeless
 Less than 1 year 29 19.3
 1–6 years 69 46.0
 7–13 years 29 19.3
 14–23 years 14 9.3
 24 to ≥30 years 9 6.0
Living arrangements (last 30 days)
 Shelter 68 45.3
 Unsheltered 34 22.7
 Institution or residential treatment facility 37 24.7
 Own/rent/apartment or house 2 1.3
 Someone else’s apartment, room, or house 6 4.0
 Car/commercial building 3 2.0
Open in a new tab

Number of years homeless ranged from less than 1 year to greater than 30 years. Nearly 46.0% of the sample had been homeless for 1 to 6 years of their life; 19.3% had been homeless less than 1 year. The remainder of the sample, 34.6% have been homeless longer than 7 years. Approximately 80.6% of our sample was homeless greater than 1 year and among that population, 54% were frail or most frail. In terms of living arrangements within the last 30 days, 45.3% lived in a shelter and 22.7% were unsheltered and lived in the streets or other outdoor areas.

In terms of substance use, the most frequent type of drug used in the last 12 months, which was perceived by the participant as causing the most serious problem, included crack/freebase (23.3%), followed by alcohol (20.7%), cocaine (7.3%), and methamphetamine (4.0%; Table 2). Alcohol was consumed by 14.7% of the sample every day. About a third (32.7%) denied drug or alcohol use.

Table 2.

Substance Use Past 12 Months (N = 150).

Characteristic n %
Self-report for drug causes the most serious problem
 None 49 32.7
 Crack/freebase 35 23.3
 Alcohol 31 20.7
 Cocaine (by itself) 11 7.3
 Methamphetamine 6 4.0
 Heroin (by itself) 5 3.3
 More than one drug causes a problem 5 3.3
 Heroin and cocaine (mixed together as speedball) 4 2.7
 Marijuana/hashish/hallucinogens 4 2.7
Types of illicit drugs used about every daya
 Alcohol 22 14.7
 Crack/freebase 16 10.5
 Cocaine 9 5.9
 Methamphetamine 2 1.3
 Marijuana 2 1.3
Current smoker (yes) 102 68.0
Open in a new tab
a

Drug use in the last 12 months.

Table 3 presents findings related to the sample mobility profile. About two thirds (67.3%) reported ability to walk independently while about one third (32.7%) reported the need to use an assistive device, such as a cane (19.3%), a walker (5.3%), or a wheelchair (2.7%). Falls within the last 30 days were reported by 22% of the population.

Table 3.

Mobility Profile (N = 150).

Characteristic n %
Walking independently
 Yes 101 67.3
 No 49 32.7
Utilization of assistive devices
 None 109 72.7
 Cane 29 19.3
 Walker 8 5.3
 Wheelchair 4 2.7
Falls in the last 30 days
 Yes 33 22.0
 No 117 78.0
Open in a new tab

Self-Reported Health Conditions

More than half (57.3%) of the participants reported that they experienced depression, followed by back pain (54.7%), hypertension (46.3%), rheumatoid arthritis (24.0%), osteoarthritis/degenerative arthritis (14.0%), asthma (12.7%), diabetes (11.3%), and bipolar/schizoaffective disorder (10.7%). Other more commonly reported conditions included heart disease (9.3%), Hepatitis C virus infection (9.3%), and anemia or other blood diseases (8.0%). In terms of health care utilization, 48% of the sample had visited an emergency room for at least one night and 46.7% had health insurance.

Descriptive Statistics for Selected Variables

Table 4 presents findings of means, standard deviations of variables. The mean days homeless was 2,413 (SD = 2,896) and the mean number for health care utilization was 7.28 (SD = 7.12). In addition, the mean resilience score was 134 (SD = 28.64) and the mean FI was 0.26 (SD = 0.15).

Table 4.

Descriptive Statistics for Selected Variables (N = 150).

Variable M SD Range
Age 52.4 6.80 40–73
Days homeless 2,413 2,896 21–14,965
Number of times homeless 13.3 12.75 1–31
Comorbidity index problem score only 3.48 2.20 0–13
Depressive symptomology (CES-D)* 25.7 12.64 4–58
Nutrition 22.10 4.41 11–30
Physical functioning score (Part 1) 57.1 31.89 0–100
Body mass index (BMI) 29.3 6.370 17–51
Activities of daily living 5.80 0.54 2–6
Texas Christian drug screen score 3.70 3.568 0–9
Health care utilization 7.28 7.12 0–35
Social support total score 54.88 20.91 19–95
Resilience scale total score 134 28.64 45–175
Frailty index 0.26 0.15 0–0.76
Open in a new tab

Note: Comorbidity index = higher scores mean higher number of conditions; CES-D = Center for Epidemiologic Scale, higher scores mean greater depressive symptoms; nutrition = higher scores mean better nutrition; physical functioning = higher scores mean poorer functioning; activities of daily living = higher scores mean better functioning; Texas Christian drug screen = higher scores mean greater drug dependency; social support = higher scores mean greater social support; resilience scale = higher scores means greater resilience; Frailty Index = higher scores mean greater frailty.

*

p < .05.

Bivariate Analysis

A Pearson (r) bivariate correlation revealed a weak relationship between frailty and being female (r = .230, p < .01). Significant moderate negative correlations were found between frailty and resilience (r = −.395, p < .01), social support (r = −.377, p < .01), and nutrition (r = −.652, p < .01). Furthermore, Spearman’s rho (rs) bivariate correlations revealed a moderate positive relationship between frailty and health care utilization (rs = .444, p < .01; Table 5).

Table 5.

Bivariate Correlations of Variables and Frailty (N = 150).

Frailty
Age .098a
Days homeless .044b
Gender .230**a
Health care utilization .444**b
Mini nutritional status (MNA) −.652**a
Texas Christian drug screen .001b
Social support −.377**a
Resilience scale −.395**a
Open in a new tab
a

Pearson correlation.

b

Spearman correlation.

*

p < .05.

**

p < .01.

Multivariate Analysis

Table 6 describes the stepwise backward linear regression among variables which have been selected based on lack of redundancy and overlap in the model. In the final model, age, gender, health care utilization, nutrition, and resilience were significantly related to frailty. At each step, the variable with the smallest nonsignificant F ratio was deleted. In Step 1, age, gender, days homeless, nutrition, drug use, health care utilization, resilience, and social support were all entered into the regression equation, F(8, 140) = 21.000, p < .001, and the multiple correlation coefficient was .545, indicating approximately 54.5% of the variance in frailty can be predicted by the variables in frailty. In the last step, length of time homeless was removed, F(5, 143) = 33.859, p < .001, and age, gender, health care utilization, nutrition, and resilience were significantly related to frailty. The squared multiple correlation coefficients was .542 explaining 54.2% of the variance in frailty can be predicted by and age, gender, health care utilization, nutrition, and resilience.

Table 6.

Results of Stepwise Backward Linear Analysis (N = 150).

Variable B SE B β p R2
Step 1
 Age .003 .001 .125 .043* .545
 Female .049 .019 .166 .012*
 Days homeless .007 .021 .021 .731
 Nutrition −.017 .002 −.498 <.001***
 Drug use .007 .024 .020 .773
 Health care utilization .086 .024 .215 .001*
 Resilience −.001 .000 −.165 .014*
 Social support .000 .000 −.059 .377
Step 2
 Age .003 .001 .120 .042* .545
 Female .046 .017 .158 .008*
 Days homeless .009 .020 .026 .657
 Nutrition −.017 .002 −.501 <.001***
 Health care utilization .087 .024 .216 <.001***
 Resilience −.001 .000 −.166 .012*
 Social support .000 .000 −.055 .398
Step 3
 Age .003 .001 .119 .042* .545
 Female .045 .017 .155 .008*
 Nutrition −.017 .002 −.499 <.001***
 Health care utilization .088 .024 .219 <.001***
 Resilience −.001 .000 −.168 .011*
 Social support .000 .000 −.057 .385
Step 4
 Age .003 .001 .125 .032* .542
 Female .045 .017 .152 .009*
 Nutrition −.017 .002 −.517 <.001***
 Health care utilization .092 .023 .229 <.001***
 Resilience −.001 .000 −.178 .006*
Open in a new tab
*

p < .05.

***

p < .001.

Discussion

The purpose of this study was to understand correlates of frailty among homeless adults. Our findings revealed that approximately 53.3% of the sample was considered frail, and 10% were considered most frail. Moreover, significant correlates of frailty included age, gender, health care utilization, and significant negative relationships were found between nutrition and resilience. We selected the FI as one out of many available frailty instruments because of its multidimensional properties; other measures may singly focus on physical (Syddall, Cooper, Martin, Briggs, & Aihie Sayer, 2003) or other functional and psychological parameters (Fried et al., 2001; Fried et al., 2004).

Chronological age was found to be predictive of frailty. The FI counts deficits that can include symptoms, signs, illnesses, and disabilities (Rockwood & Mitnitski, 2011); thus, it is reasonable that as the population ages, chronic diseases will similarly increase. Published literature likewise suggests that as age increases, frailty levels similarly increase (Crews & Zavotka, 2006; Song, Mitnitski, & Rockwood, 2010; Yu et al., 2012). Currently, nearly one third of chronically homeless adults are above the age of 55 (LAHSA, 2011). As the homeless population continues to age, it will become increasingly necessary to focus on translational research which will address frailty.

Gender was also significantly related to frailty; more specifically, women were found to be frailer than men. In fact, regardless of type of frailty measure used, data reveal that women are at greater risk for frailty when compared with men (Goggins et al., 2005; Woo et al., 2005), live a greater number of years, and may have greater functional limitations (Graham et al., 2009). This area warrants further study as it relates to homeless women.

Our data further suggest that increased health care utilization was significantly related to increasing levels of frailty. Among homeless populations, health care utilization is a significant issue as many homeless adults utilize emergency departments frequently (Hahn et al., 2006; Kushel, Perry, Bangsberg, Clark, & Moss, 2002). One study examining factors of emergency department utilization among homeless adults (N = 2,578) found that 40.4% had one or more emergency department visits (Kushel et al., 2002). In our sample, 48% had visited the emergency department for at least one night, about one in five self-reported having an illness as a contributing factor to becoming homeless, and about half did not have health insurance. This area necessitates further exploration as it has significant health care cost and quality of life implications.

Among homeless populations, a greater degree of chronic disease burden may be one of the contributing factors relating to increased emergency department utilization (Sadowski, Kee, VanderWeele, & Buchanan, 2009). The most common conditions reported by our participants included depression, followed by back pain, hypertension, rheumatoid arthritis, and osteoarthritis/degenerative arthritis. We know that chronic health conditions are frequently reported by homeless populations (Garibaldi et al., 2005; Wiersma et al., 2010). In fact, one study found that nearly 85% of homeless adults older than 50 years of age had at least one chronic condition and the top three conditions were hypertension, arthritis/musculoskeletal disorders, and psychiatric conditions (Garibaldi et al., 2005).

We also found that poorer nutrition scores were related to higher frailty scores. Authors acknowledge that one of the key issues related to frailty is an impairment of nutritional status, which is a critical issue among older adults (Bartali et al., 2006; Fulop et al., 2010; Kaiser, Bandinelli, & Lunenfeld, 2010), as well as homeless adults. In a secondary study among homeless adults in this sample, data revealed that 25% were food insufficient and not having adequate quantity to eat (Baggett et al., 2011). For some homeless adults, scavenging dumpsters, stealing food, and pawning personal belongings (Richards & Smith, 2006) were means by which to obtain food. Although some measures of frailty may include nutrition as a component, due to the condition of homelessness and challenges with ascertaining adequate nutrition, we decided to include this variable as an antecedent. Future research should focus on identifying nutritional deficiencies and intervening with supplementation and measuring effectiveness among this population.

This study yielded several variables that were not associated with frailty. First, length of time homeless was also not related to frailty; to our knowledge, no previous research has investigated the relationship between frailty and length of time homeless. Approximately 80.6% of our sample was homeless more than 1 year and among that population 54% were frail and most frail. Due to the cross-sectional nature of the study, it is difficult to determine whether these findings would be consistent with a longitudinal study. Due to the lack of variability with length of time homeless measure, it is difficult to determine the role that this measure had on frailty. However, it seems logical that homeless persons should have higher degrees of frailty especially due to the constellation of physical, psychological, and social challenges. Moreover, transition between frailty states may be improved if resources became more accessible.

No relationships were noted with respect to frailty and either drug use or social support. Previous studies in the general population have found that social vulnerability is related to frailty (Bilotta et al., 2010). It is plausible that in this sample participants perceive support rather than receive support. Among older adults (N = 2,032), decreased social support was related to increasing frailty; furthermore, one strategy to delaying onset of frailty is active participation in the community (Woo et al., 2005). In a study of multidimensional social support among homeless adults (N = 544), Hwang et al. (2009) found that the sample included had high levels of emotional and instrumental social support. Our findings may be related to the fact that while this population self-reports high levels of emotional/information support, tangible support, affectionate support, or positive social interaction; they may not actually receive support. This area necessitates further exploration. Data suggest that substance use is a significant issue among homeless populations (Christiani, Hudson, Nyamathi, Mutere, & Sweat, 2008; A. Nyamathi, Branson, et al., 2012); however, substance use was not found to be correlated with frailty in the multiple regression analyses. It is possible that self-reported drug use was not accurate. It is also the case that a portion of the homeless sample was living in a substance abuse treatment facility; thus, substance abuse may have been under control for some participants.

Frailty is a public health challenge; homeless populations have a greater burden of disease and challenges abound in terms of comorbid conditions, physical functioning, and nutrition. These study findings raise several research questions and encourage us to contemplate different models of care. First, it may be necessary that service agencies have frontline geriatric nursing triage to accurately case manage clients, especially those who need higher acuity care. It may also be necessary to develop shelter-based convalescence (van Laere, de Wit, & Klazinga, 2009) facilities for homeless populations. In addition to nurse case management, another promising component of an intervention program may be to utilize a chronic disease self-management program (CDSMP), initially established for helping patients manage arthritis (K. R. Lorig & Holman, 2003; K. Lorig, Laurin, & Holman, 1984; K. Lorig, Ritter, & Plant, 2005).

It may also be beneficial for service agencies to provide health promotion activities in day centers and to have nutritionists on staff guiding and planning meals. Among an already vulnerable population, it remains to be seen whether utilizing a FI would enable a clearer identification of issues and subsequent rendering of services or care. Taken together, it will be critical that clinicians and service providers work together and begin to consider clinical applications of frailty screening. It may also be necessary to triangulate services focusing on nutrition, exercise, and case management of chronic disease processes in an effort to care for the most vulnerable. In essence, these findings serve as an impetus and a foundation for understanding frailty that will undoubtedly inform future nurse-led larger replicative studies that will lead to interventions.

A discussion of limitations is warranted; first, the study is limited to homeless men and women between 40 and 73 years of age and cannot be generalized to other populations. Furthermore, this study only utilized one frailty instrument (e.g., FI). Future research should compare and contrast frailty instruments among this population. Similarly, cross-sectional studies cannot infer directional relationships or causality. In fact, due to the nature of this design, it is not possible that we are able to completely understand the nature of the relationship between longer time spent on the street and frailty. Furthermore, self-report data are prone to errors and may cause bias in the data. Future research should focus on replicating this study with a larger sample size, without a minimum inclusion age in an effort to obtain greater variability with the measures.

Acknowledgments

We acknowledge the support of Dr. Arnold Mitnitski and Dr. Kenneth Rockwood for providing guidance with utilizing the frailty index. We similarly thank the Los Angeles–based skid row homeless service agencies and homeless participants who generously gave of their time, experiences, and resources.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institute of Health (NIH)/Nursing Research (NINR) T32 NR007077 and the University of California Los Angeles (UCLA) Dissertation Year Fellowship Award.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

  1. Aranda MP, Ray LA, Snih SA, Ottenbacher KJ, Markides KS. The protective effect of neighborhood composition on increasing frailty among older Mexican Americans: A barrio advantage? Journal of Aging and Health. 2011;23:1189–1217. doi: 10.1177/0898264311421961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arranz L, de Vicente A, Munoz M, De la Fuente M. Impaired immune function in a homeless population with stress-related disorders. Neuroimmunomodulation. 2009;16:251–260. doi: 10.1159/000212386. [DOI] [PubMed] [Google Scholar]
  3. Baggett TP, Singer DE, Rao SR, O’Connell JJ, Bharel M, Rigotti NA. Food insufficiency and health services utilization in a national sample of homeless adults. Journal of General Internal Medicine. 2011;26:627–634. doi: 10.1007/s11606-011-1638-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bartali B, Frongillo EA, Bandinelli S, Lauretani F, Semba RD, Fried LP, Ferrucci L. Low nutrient intake is an essential component of frailty in older persons. Journals of Gerontology: Series A, Biological Sciences and Medical Sciences. 2006;61:589–593. doi: 10.1093/gerona/61.6.589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bergman H, Beland F, Karunananthan S, Hummel S, Hogan D, Wolfson C. Developing a working framework for understanding frailty. Gerontologie et societe. 2004;109:15–29. [Google Scholar]
  6. Bergman H, Ferrucci L, Guralnik J, Hogan DB, Hummel S, Karunananthan S, Wolfson C. Frailty: An emerging research and clinical paradigm—Issues and controversies. Journals of Gerontology: Series A, Biological Sciences and Medical Sciences. 2007;62:731–737. doi: 10.1093/gerona/62.7.731. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bilotta C, Case A, Nicolini P, Mauri S, Castelli M, Vergani C. Social vulnerability, mental health and correlates of frailty in older outpatients living alone in the community in Italy. Aging & Mental Health. 2010;14:1024–1036. doi: 10.1080/13607863.2010.508772. [DOI] [PubMed] [Google Scholar]
  8. Brown RT, Kiely DK, Bharel M, Mitchell SL. Geriatric syndromes in older homeless adults. Journal of General Internal Medicine. 2012;27:16–22. doi: 10.1007/s11606-011-1848-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Christiani A, Hudson AL, Nyamathi A, Mutere M, Sweat J. Attitudes of homeless and drug-using youth regarding barriers and facilitators in delivery of quality and culturally sensitive health care. Journal of Child and Adolescent Psychiatric Nursing. 2008;21:154–163. doi: 10.1111/j.1744-6171.2008.00139.x. [DOI] [PubMed] [Google Scholar]
  10. Crews DE, Zavotka S. Aging, disability, and frailty: Implications for universal design. Journal of Physiological Anthropology. 2006;25:113–118. [PubMed] [Google Scholar]
  11. Dachner N, Tarasuk V. Homeless “squeegee kids”: Food insecurity and daily survival. Social Science & Medicine. 2002;54:1039–1049. doi: 10.1016/s0277-9536(01)00079-x. [DOI] [PubMed] [Google Scholar]
  12. DiMaria-Ghalili R, Guenter P. Mini nutritional assessment—How to try this. 2008 doi: 10.1097/01.NAJ.0000308962.37976.c9. Retrieved from http://www.nursingcenter.com/_PDF_.aspx?an=00000446-200802000-00030. [DOI] [PubMed]
  13. Epel ES, Blackburn EH, Lin J, Dhabhar FS, Adler NE, Morrow JD, Cawthon RM. Accelerated telomere shortening in response to life stress. Proceedings of the National Academy of Sciences of the United States of America. 2004;101:17312–17315. doi: 10.1073/pnas.0407162101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fazel S, Khosla V, Doll H, Geddes J. The prevalence of mental disorders among the homeless in western countries: Systematic review and meta-regression analysis. PLoS Medicine. 2008;5:e225. doi: 10.1371/journal.pmed.0050225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Flaskerud JH, Winslow BJ. Conceptualizing vulnerable populations health-related research. Nursing Research. 1998;47:69–78. doi: 10.1097/00006199-199803000-00005. [DOI] [PubMed] [Google Scholar]
  16. Fountain J, Howes S, Marsden J, Taylor C, Strang J. Drug and alcohol use and the link with homelessness: Results from a survey of homeless people in London. Addiction Research & Theory. 2003;11:245–256. doi: 10.1080/1606635031000135631. [DOI] [Google Scholar]
  17. Fried LP, Ferrucci L, Darer J, Williamson JD, Anderson G. Untangling the concepts of disability, frailty, and comorbidity: Implications for improved targeting and care. Journals of Gerontology: Series A, Biological Sciences and Medical Sciences. 2004;59:255–263. doi: 10.1093/gerona/59.3.m255. [DOI] [PubMed] [Google Scholar]
  18. Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J Cardiovascular Health Study Collaborative Research Group. Frailty in older adults: Evidence for a phenotype. Journals of Gerontology: Series A, Biological Sciences and Medical Sciences. 2001;56:M146–M156. doi: 10.1093/gerona/56.3.m146. [DOI] [PubMed] [Google Scholar]
  19. Fried LP, Walston J. Frailty and failure to thrive. In: Hazzard WR, Blass JP, Halter JB, Ouslander JG, Tinetti ME, editors. Principles of geriatric medicine and gerontology. 5. Vol. 5. New York, NY: McGraw-Hill; 2003. pp. 1487–1502. [Google Scholar]
  20. Fulop T, Larbi A, Witkowski JM, McElhaney J, Loeb M, Mitnitski A, Pawelec G. Aging, frailty and age-related diseases. Biogerontology. 2010;11:547–563. doi: 10.1007/s10522-010-9287-2. [DOI] [PubMed] [Google Scholar]
  21. Garibaldi B, Conde-Martel A, O’Toole TP. Self-reported comorbidities, perceived needs, and sources for usual care for older and younger homeless adults. Journal of General Internal Medicine. 2005;20:726–730. doi: 10.1111/j.1525-1497.2005.0142.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Garre-Olmo J, Calvo-Perxas L, Lopez-Pousa S, de Gracia Blanco M, Vilalta-Franch J. Prevalence of frailty phenotypes and risk of mortality in a community-dwelling elderly cohort. Age & Ageing. 2013;42:46–51. doi: 10.1093/ageing/afs047. [DOI] [PubMed] [Google Scholar]
  23. Gelberg L, Linn L, Mayer-Oakes S. Differences in health status between older and younger homeless adults. Journal of the American Geriatrics Society. 1990;38:1220–1229. doi: 10.1111/j.1532-5415.1990.tb01503.x. [DOI] [PubMed] [Google Scholar]
  24. Geronimus AT, Hicken MT, Keene D, Bound J. “Weathering” and age patterns of allostatic load scores among blacks and whites in the United States. American Journal of Public Health. 2006;96:826–833. doi: 10.2105/AJPH.2004.060749. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Geronimus AT, Hicken MT, Pearson JA, Seashols SJ, Brown KL, Cruz TD. Do US black women experience stress-related accelerated biological aging? A novel theory and first population-based test of black-white differences in telomere length. Human Nature. 2010;21:19–38. doi: 10.1007/s12110-010-9078-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Gielen E, Verschueren S, O’Neill TW, Pye SR, O’Connell MD, Lee DM, Boonen S. Musculoskeletal frailty: A geriatric syndrome at the core of fracture occurrence in older age. Calcified Tissue International. 2012;91:161–177. doi: 10.1007/s00223-012-9622-5. [DOI] [PubMed] [Google Scholar]
  27. Gobbens RJ, van Assen MA, Luijkx KG, Schols JM. Testing an integral conceptual model of frailty. Journal of Advanced Nursing. 2012;68:2047–2060. doi: 10.1111/j.1365-2648.2011.05896.x. [DOI] [PubMed] [Google Scholar]
  28. Goggins WB, Woo J, Sham A, Ho SC. Frailty index as a measure of biological age in a Chinese population. Journals of Gerontology: Series A, Biological Sciences and Medical Sciences. 2005;60:1046–1051. doi: 10.1093/gerona/60.8.1046. [DOI] [PubMed] [Google Scholar]
  29. Graham JE, Snih SA, Berges IM, Ray LA, Markides KS, Ottenbacher KJ. Frailty and 10-year mortality in community-living Mexican American older adults. Gerontology. 2009;55:644–651. doi: 10.1159/000235653. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Greenberg GA, Rosenheck RA. Jail incarceration, homelessness, and mental health: A national study. Psychiatric Services. 2008;59:170–177. doi: 10.1176/appi.ps.59.2.170. [DOI] [PubMed] [Google Scholar]
  31. Hahn JA, Kushel MB, Bangsberg DR, Riley E, Moss AR. Brief report: The aging of the homeless population: Fourteen-year trends in San Francisco. Journal of General Internal Medicine. 2006;21:775–778. doi: 10.1111/j.1525-1497.2006.00493.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Hamilton AB, Poza I, Washington DL. Homelessness and trauma go hand-in-hand: Pathways to homelessness among women veterans. Women’s Health Issues. 2011;21(Suppl 4):S203–S209. doi: 10.1016/j.whi.2011.04.005. [DOI] [PubMed] [Google Scholar]
  33. Hwang SW, Kirst MJ, Chiu S, Tolomiczenko G, Kiss A, Cowan L, Levinson W. Multidimensional social support and the health of homeless individuals. Journal of Urban Health. 2009;86:791–803. doi: 10.1007/s11524-009-9388-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Kaiser M, Bandinelli S, Lunenfeld B. Frailty and the role of nutrition in older people. A review of the current literature. Acta Biomedica: Atenei Parmensis. 2010;81(Suppl 1):37–45. [PubMed] [Google Scholar]
  35. Katz S, Akpom CA. Index of activities of daily living (ADL) Medical Care. 1976;14(Suppl 5):116–118. doi: 10.1097/00005650-197605001-00018. [DOI] [PubMed] [Google Scholar]
  36. Knight K, Simpson DD, Hiller ML. Screening and referral for substance-abuse treatment in the criminal justice system. In: Leukefeld CG, Tims F, Farabee D, editors. Treatment of drug offenders: Policies and issues. New York, NY: Springer; 2002. pp. 259–272. [Google Scholar]
  37. Kotrschal A, Ilmonen P, Penn DJ. Stress impacts telomere dynamics. Biology Letters. 2007;3:128–130. doi: 10.1098/rsbl.2006.0594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Kushel MB, Perry S, Bangsberg D, Clark R, Moss AR. Emergency department use among the homeless and marginally housed: Results from a community-based study. American Journal of Public Health. 2002;92:778–784. doi: 10.2105/ajph.92.5.778. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Lorig K, Laurin J, Holman HR. Arthritis self-management: A study of the effectiveness of patient education for the elderly. The Gerontologist. 1984;24:455–457. doi: 10.1093/geront/24.5.455. [DOI] [PubMed] [Google Scholar]
  40. Lorig KR, Holman H. Self-management education: History, definition, outcomes, and mechanisms. Annals of Behavioral Medicine. 2003;26:1–7. doi: 10.1207/S15324796ABM2601_01. [DOI] [PubMed] [Google Scholar]
  41. Lorig K, Ritter PL, Plant K. A disease-specific self-help program compared with a generalized chronic disease self-help program for arthritis patients. Arthritis & Rheumatism. 2005;53:950–957. doi: 10.1002/art.21604. [DOI] [PubMed] [Google Scholar]
  42. Los Angeles Housing Services Administration. Greater Los Angeles homeless count. Los Angeles, CA: Author; 2011. [Google Scholar]
  43. McDowell I. Measuring health: A guide to rating scales and questionnaires. Oxford, UK: Oxford University Press; 2006. [Google Scholar]
  44. Mitnitski AB, Mogilner AJ, MacKnight C, Rockwood K. The accumulation of deficits with age and possible invariants of aging. Scientific World Journal. 2002;2:1816–1822. doi: 10.1100/tsw.2002.861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Nyamathi A, Branson C, Kennedy B, Salem B, Khalilifard F, Marfisee M, Leake B. Impact of nursing intervention on decreasing substances among homeless youth. American Journal on Addictions. 2012;21:558–565. doi: 10.1111/j.1521-0391.2012.00288.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Nyamathi A, Hudson A, Greengold B, Leake B. Characteristics of homeless youth who use cocaine and methamphetamine. American Journal on Addictions. 2012;21:243–249. doi: 10.1111/j.1521-0391.2012.00233.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Nyamathi A, Leake B, Albarran C, Zhang S, Hall E, Farabee D, Faucette M. Correlates of depressive symptoms among homeless men on parole. Issues in Mental Health Nursing. 2011;32:501–511. doi: 10.3109/01612840.2011.569111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Nyamathi AM, Leake B, Gelberg L. Sheltered versus nonsheltered homeless women differences in health, behavior, victimization, and utilization of care. Journal of General Internal Medicine. 2000;15:565–572. doi: 10.1046/j.1525-1497.2000.07007.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Puts ME, Lips P, Ribbe M, Deeg DH. The effect of frailty on residential/nursing home admission in the Netherlands independent of chronic diseases and functional limitations. European Journal of Ageing. 2005;2:264–274. doi: 10.1007/s10433-005-0011-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Radloff LS. The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401. [Google Scholar]
  51. Richards R, Smith C. The impact of homeless shelters on food access and choice among homeless families in Minnesota. Journal of Nutrition Education and Behavior. 2006;38:96–105. doi: 10.1016/j.jneb.2005.11.031. [DOI] [PubMed] [Google Scholar]
  52. Rockwood K, Mitnitski A. Frailty in relation to the accumulation of deficits. Journals of Gerontology: Series A, Biological Sciences and Medical Sciences. 2007;62:722–727. doi: 10.1093/gerona/62.7.722. [DOI] [PubMed] [Google Scholar]
  53. Rockwood K, Mitnitski A. Frailty defined by deficit accumulation and geriatric medicine defined by frailty. Clinics in Geriatric Medicine. 2011;27:17–26. doi: 10.1016/j.cger.2010.08.008. [DOI] [PubMed] [Google Scholar]
  54. Rockwood K, Song X, Mitnitski A. Changes in relative fitness and frailty across the adult lifespan: Evidence from the Canadian national population health survey. Canadian Medical Association Journal. 2011;183:E487–E494. doi: 10.1503/cmaj.101271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Sadowski LS, Kee RA, VanderWeele TJ, Buchanan D. Effect of a housing and case management program on emergency department visits and hospitalizations among chronically ill homeless adults: A randomized trial. Journal of the American Medical Association. 2009;301:1771–1778. doi: 10.1001/jama.2009.561. [DOI] [PubMed] [Google Scholar]
  56. Sangha O, Stucki G, Liang MH, Fossel AH, Katz JN. The self-administered comorbidity questionnaire: A new method to assess comorbidity for clinical and health services research. Arthritis & Rheumatism. 2003;49:156–163. doi: 10.1002/art.10993. [DOI] [PubMed] [Google Scholar]
  57. Searle SD, Mitnitski A, Gahbauer EA, Gill TM, Rockwood K. A standard procedure for creating a frailty index. BMC Geriatrics. 2008;8:24. doi: 10.1186/1471-2318-8-24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Sermons W, Henry M. Demographics of homelessness series: The rising elderly population. 2010 Retrieved from http://www.issuelab.org/research/demographics_of_homelessness_series_the_rising_elderly_population.
  59. Sherbourne C, Stewart A. The medical outcomes social (MOS) survey. Social Science & Medicine. 1991;32:705–714. doi: 10.1016/0277-9536(91)90150-b. [DOI] [PubMed] [Google Scholar]
  60. Song X, Mitnitski A, Rockwood K. Prevalence and 10-year outcomes of frailty in older adults in relation to deficit accumulation. Journal of the American Geriatrics Society. 2010;58:681–687. doi: 10.1111/j.1532-5415.2010.02764.x. [DOI] [PubMed] [Google Scholar]
  61. Stein JA, Andersen RM, Robertson M, Gelberg L. Impact of hepatitis B and C infection on health services utilization in homeless adults: A test of the Gelberg-Andersen behavioral model for vulnerable populations. Health Psychology. 2012;31:20–30. doi: 10.1037/a0023643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Syddall H, Cooper C, Martin F, Briggs R, Aihie Sayer A. Is grip strength a useful single marker of frailty? Age & Ageing. 2003;32:650–656. doi: 10.1093/ageing/afg111. [DOI] [PubMed] [Google Scholar]
  63. Syddall H, Roberts HC, Evandrou M, Cooper C, Bergman H, Aihie Sayer A. Prevalence and correlates of frailty among community-dwelling older men and women: Findings from the Hertfordshire cohort study. Age & Ageing. 2010;39:197–203. doi: 10.1093/ageing/afp204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. U.S. Department of Housing and Urban Development. Review of Stewart B. McKinney homeless programs administered by HUD. Washington, DC: Author; 1995. [Google Scholar]
  65. van Laere I, de Wit M, Klazinga N. Shelter-based convalescence for homeless adults in Amsterdam: A descriptive study. BMC Health Services Research. 2009;9:208. doi: 10.1186/1472-6963-9-208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Vellas B, Guigoz Y, Garry PJ, Nourhashemi F, Bennahum D, Lauque S, Albarede JL. The mini nutritional assessment (MNA) and its use in grading the nutritional state of elderly patients. Nutrition. 1999;15:116–122. doi: 10.1016/s0899-9007(98)00171-3. [DOI] [PubMed] [Google Scholar]
  67. Wagnild GM. A review of the resilience scale. Journal of Nursing Measurement. 2009;17:105–113. doi: 10.1891/1061-3749.17.2.105. [DOI] [PubMed] [Google Scholar]
  68. Wagnild GM, Young HM. Development and psychometric evaluation of the resilience scale. Journal of Nursing Measurement. 1993;1:165–178. [PubMed] [Google Scholar]
  69. Weissman MM, Sholomskas D, Pottenger M, Prusoff BA, Locke BZ. Assessing depressive symptoms in five psychiatric populations: A validation study. American Journal of Epidemiology. 1977;106:203–214. doi: 10.1093/oxfordjournals.aje.a112455. [DOI] [PubMed] [Google Scholar]
  70. Wiersma P, Epperson S, Terp S, Lacourse S, Finton B, Drenzek C, Finelli L. Episodic illness, chronic disease, and health care use among homeless persons in metropolitan Atlanta, Georgia, 2007. Southern Medical Journal. 2010;103:18–24. doi: 10.1097/SMJ.0b013e3181c46f79. [DOI] [PubMed] [Google Scholar]
  71. Woo J, Goggins W, Sham A, Ho SC. Social determinants of frailty. Gerontology. 2005;51:402–408. doi: 10.1159/000088705. [DOI] [PubMed] [Google Scholar]
  72. Yu P, Song X, Shi J, Mitnitski A, Tang Z, Fang X, Rockwood K. Frailty and survival of older Chinese adults in urban and rural areas: Results from the Beijing longitudinal study of aging. Archives of Gerontology and Geriatrics. 2012;54:3–8. doi: 10.1016/j.archger.2011.04.020. [DOI] [PubMed] [Google Scholar]

RESOURCES