Describing the health of the absolutely homeless population in downtown Calgary 2008.

Nicholson, Cherie L. ; Graham, John R. ; Emery, J.C. Herbert 等

Abstract

People experiencing homelessness carry a heavy burden of illness and disease compared to those in the general population. We analysed data from the Calgary 2008 Vulnerability Index (VI) survey to assess the health of individuals experiencing absolute homelessness. In October and December 2008 a convenience sample of 137 homeless individuals were surveyed with a variant of the Common Ground (New York) VI instrument. Calgary's homeless population has alarmingly poor health. The Calgary homeless have higher reported rates of liver disease, kidney disease, and frostbite when compared with other North American VI survey locations. Homeless females in Calgary are less prevalent than males but their health is much poorer. Homeless individuals in Calgary are sick and at an increased risk of dying prematurely. The VI is an effective tool to assess the health of the homeless population and should be further utilized to focus immediate interventions.

Keywords: homeless persons, health measures, physical health, mental health, addictions

Resume

Les sans-abris ont generalement un taux de maladie et d'infection plus eleve que la population en generale. Notre etude est base sur l'analyse des donnees du recensement de l'Indice de Vulnerabilite de Calgary (2008). Notre objectifest d'evaluer l'eat de sante des individus qui sont entierement sans-abris. L'etude est basee sur un echantillon de 137 sans-abris a partir d'une variation de la methode Common Ground V1 (New York). L'on constate une evolution alarmante de l'etat de sante de la population des sans-abris a Calgary. Ces sans-abris ont un taux plus eleve de maladie renale, du foie et de gelure comparativement aux autres recensements de type V1 de differentes villes nord-americaines. Les femmes sans-abris sont moins nombreuses que les hommes. Toutefois, leurs etat de sante est encore plus alarmant que les males. On constate donc que l'etat de sante des sans-abris est precaire et le risque d'une mort prematuree est fort eleve au sein de cette population. Common Ground V1 est un outil d'etude efficace pour mesurer l'etat de sante de la population des sans-abris afin de mettre l'accent sur des interventions immediates pour aider la population des sans-abris.

Mots cles: sans-abris, indices de sante, sante physique, sante mentale, dependance

Introduction

Absolute homelessness refers to those without physical shelter and includes those who sleep outdoors, in emergency shelters, vehicles, or other places unfit for human habitation (City of Calgary 2008). Those experiencing absolute homelessness are at a high risk for developing serious, chronic health conditions. People living on the streets, also known as rough sleepers, and in homeless shelters are exposed to and develop a wide range of opportunistic infections and diseases, as well as suffer from a number of chronic health conditions which are difficult to manage without adequate shelter and food. Poor health and inadequate health management result in mortality rates in far greater numbers than the general population (Barrow et al. 1999; Hwang 2000; O'Connell 2005; Schanzer et al. 2007). According to a 1998 study in Boston, the strongest risk factors for death among homeless adults were HIV! AIDS, renal disease, cold-related injury, liver disease, arrhythmia, and history of substance abuse involving injection drugs or alcohol (Hwang et al. 1998). The high incidence of mental health and addictions problems further exacerbates efforts to identify, treat, and control the spread of infectious diseases and properly manage health conditions which also result in higher mortality rates.

Recent direction in housing vulnerable people has turned to a "housing first" (Padgett, Gulcur and Tsemberis 2006) approach. This leads to the logical assumption that health interventions for homeless persons should start with immediate re-housing (housing first), in order to reduce the greatest risk factor for continuing ill health; that being the stress of living on the streets. Subsequently, issues of health care, substance use, and other conditions can be more effectively managed. Given the large numbers of homeless persons in any given locality, deciding on housing priorities for those most in need of shelter should be a primary consideration. Housing those most at risk, those severely impacted by high risk diseases and infections, and those with multiple diagnoses, including physical and mental health and addictions problems, has been a recent approach in a number of North American urban areas such as Toronto, Boston, Portland, and Calgary. This type of initiative requires the implementation of screening methods that determine those at greatest health risk. Addressing the health of the homeless through re-housing efforts, improving access to health services (Gelberg, Andersen and Leake 2000; Kushel, Vittinghoff and Haas 2001; McCormack and Macintosh 2001; Nyamathi, Leake and Gelberg 2000), and preventative interventions (Kushel et al. 2002; Weinreb et al. 1998) requires information on personal health histories (D'Amore et al. 2001) and the severity of health related issues (Kessell et al. 2006). Knowing who the homeless are, where they are living, and what their socio-demographic, health, housing, and systems interaction profiles are, can significantly increase capacity to prioritize those who are most vulnerable and in greatest need of immediate re-housing.

Background

The plight of health challenges, so eloquently described in Dickcnsian novels such as Oliver Twist were largely ignored by housing researchers until about twenty years ago. Since then numerous articles have explored many facets of this issue. The following is a sample of more than 300 articles over the last 20 years that deal with various aspects of health and homelessness. Early efforts linked homelessness and health status (Caton 1990; Schutt and Garrett 1988), and were soon followed by those that examined sub-populations of homeless persons. With the link between mental health and homelessness clearly established, subsequent work with this population turned towards inclusion of additional co-morbid disorders such as substance abuse and physical health problems (Aday 1994; Bebout 1999; Sosin and Bruni 1997). Concurrently some investigators explored homelessness and health status among various populations: including youth (Chen et al. 2006; Gelberg, Linn and Mayer-Oakes 1990; geuler 1991) and women (Ritchey, la Gory and Mullis 1991). HIV/AIDS in homeless populations was also included in this initial look at health vulnerabilities (Metsch et al. 1995). Mortality rates and risk of mortality among homeless persons for various disorders have been well documented within the literature (Cheung and Hwang 2004; Hwang 2001; Kidder et al. 2007; Morrison 2009).

Health implications for homeless persons with other disorders and diseases have, more recently, also received considerable attention. Of significant interest to shelter managers and health care personnel are highly transmittable diseases such as tuberculosis (Badiaga et al. 2009; de Vries and van Hest 2006), hepatitis-C (Nyamathi et al. 2006; Sherriff and Mayon-White 2003), respiratory disease (Badiaga et al. 2009), and cardiovascular disease (Kim et al. 2008). Most of these investigations concentrated on the identification of the prevalence of disorders.

Access to health care for vulnerable people has also received attention, especially in the United States (U.S.) where lack of universal coverage makes access more problematic (Nickasch and Marnocha 2009). The evidence from these studies clearly suggests that homeless persons have more difficulty accessing health care, especially if they had pre-existing substance abuse or mental health conditions (Savage and Lindsell 2008). Intervention efforts have primarily concentrated on treating health conditions experienced by homeless people while they are still homeless (Lashley 2007; Lewis, Andersen and Gelberg 2003; Martins 2008). An effort to re-house homeless people as part of a health intervention strategy appears to be gradually developing. This is evidenced by housing first and Streets to Homes strategies in various cities such as Boston, Portland, Oregon, and Toronto. As a result of the wide gap between available housing for homeless people (usually rent subsidized), many of whom also require support services to stay housed, and available housing units and support staff, some housing programs have attempted to develop a method of prioritizing housing need: those with dual diagnoses (mental health and addictions) or chronic, serious medical conditions are prioritized at the top of the list.

In recognition of the role that health risks play in the lives of homeless people, in 2007, Common Ground, a New York City agency that develops innovative housing programs to end homelessness, created the Vulnerability Index (VI) for use in its Streets to Home program as a tool for assessing the health of people experiencing homelessness and determining their need for immediate housing. Common Ground uses the VI to identify and prioritize members of the homeless population for housing according to the number of serious risk factors a person has. The instrument collects self-reported information regarding demographics, homeless history, social history, health conditions, income, employment, education, and ethnic background (Common Ground 2007). Health factors that are rated based on perceived seriousness of illness and the length of homelessness together constitutes a vulnerability rating. To date, the index has been used in nine North American cities to assess the health of the homeless living on the street and in shelters (Common Ground 2008; Common Ground Institute for Portland's Bureau of Housing and Community Development n.d.; Common Ground Institute for Park Center, Inc and Metropolitan Homelessness Commission n.d.; LA Health Action n.d.; Unity of Greater New Orleans n.d.). Data from the implementation of the VI in multiple cities are being used to assess the social costs arising from the frequent use of publicly funded health and corrections services. This costing information, in addition to the poor condition of participants' health status, serves as a powerful resource to mobilize governments and communities to address homelessness.

The present study describes the health of the homeless population in Calgary, Alberta using the VI with 132 homeless people. Profiles are then compared with those from American cities. The survey was conducted in October and December 2008 by the Calgary Homeless Foundation (CHF), an organization with the mandate to end homelessness in Calgary by 2018. The CHF modified the VI to the Calgary and Canadian context. The study was approved by the Conjoint Faculties Research Ethics Board at the University of Calgary.

Methods

Instrument: the Vulnerability Index

Development of the VI by Common Ground relied heavily on previous work by O'Connell (2005) working at Boston's Healthcare for the Homeless Program (BHCHP) who identified and described specific health conditions that place homeless individuals at an increased risk of dying on the street. Based on work by O'Connell and Hwang (Cheung and Hwang, 2004; Hwang 2001; Hwang et al. 1998), Levitt et al. (2009) developed and pilot tested the first version of a VI. The instrument was further modified using the experiences of staff in the Common Ground Program (Kanis 2009). Risk factors include serious medical conditions, health care utilization, and length of time homeless. According to Common Ground:

For individuals who have been homeless for at least six months, one or more of the following markers place them at heightened risk of mortality: 1) more than three hospitalizations or emergency room visits in a year; 2) more than three emergency room visits in the previous three months; 3) aged 60 or older; 4) cirrhosis of the liver; 5) end-stage renal disease; 6) history of frostbite, immersion foot, or hypothermia; and 7) HIV+/AIDS (cited in Kanis 2009, p. 1).

The VI instrument can be used to assess the health needs of people experiencing homelessness to identify and prioritize the most vulnerable individuals on the street for re-housing. Individuals who have been homeless for more than six consecutive months and have at least one of the indicators listed above are classified as vulnerable. Those who have a higher number of these indicators are considered to be more vulnerable and have a higher risk of mortality. Survey participants are ranked based on their vulnerability which then provides criteria for prioritizing individuals for housing. Although used by homeless service providers in diverse locations, primarily in the U.S., the psychometric properties of this instrument are not reported in the literature. Nonetheless, it has strong face validity and has been accepted by some service recipients as reflecting their status and needs (Kanis 2009).

The VI was modified for the Calgary and Canadian context by New York's Common Ground Innovation Team in cooperation with the CHE The modification involved changing some terms in the survey to reflect Canadian geography and relevant terms; for example, the Canadian Armed Forces and Canadian provinces. The modified survey included 30 questions relating to demographic information, housing status, homelessness history, health conditions, social history, employment, income, ethnicity, and citizenship. The majority of the questions required a yes or no response (allowing for non-response or unknown). Two questions pertaining to substance abuse and mental illness required observational input from the interviewer. For health conditions, participants were asked if they currently or previously had, or were told they had by a healthcare provider, the listed medical condition. The CHF worked in partnership with local agencies to administer the survey.

Participants

The VI was administered in 2008 as a street-level survey. In October and December five teams of four trained volunteers were sent out on three consecutive nights between the hours of 3:00 a.m. and 6:00 a.m. to specific areas in the city to look for people sleeping outside. Locations chosen were known to be locales frequented by homeless individuals. Participants were also sampled from two local shelters. A convenience sampling methodology was used. When an individual was observed he or she was approached by the team and was asked to provide written consent to participate. The participants were men and women aged 18 years and over who reported absolute homelessness, and were on the streets or in a shelter on one of the nights that the survey was conducted. The survey took approximately 30 minutes to complete. Upon completion of the survey the participant received a $5 (CAD) gift card for a nationally franchised coffee shop. Fifty-seven individuals completed the October survey when survey administration focused on the area surrounding two downtown shelters and one downtown neighborhood. In December, 80 individuals completed the survey when it was administered over a larger geographic area including the downtown core, one community each in the south and north of the downtown core, and an area which separates the downtown area from residential neighbourhoods.

Analysis

Analysis of the data was conducted using Stata/IC 10.0 (StataCorp 2008). Descriptive statistics including frequency counts and proportions were calculated for each health condition for data pooled from the two survey dates. The sample was stratified by Aboriginal status (First Nations and Metis versus non-Aboriginal) and sex. Individuals were identified by a code using letters of their last name and date of birth. Duplicates (n=5) were identified through cross referencing and comparison with the completed instruments and subsequently removed from the data set prior to analysis.

Sample frequencies and proportions are compared, where possible, with results from five other U.S. VI survey locations: Nashville, Los Angeles, Santa Monica, Portland, and New Orleans. All of these surveys had been done in the previous two years using the similarly focused questions and like methods of survey administration. It is not clear if those who sleep rough have a different profile of physical health conditions, mental health, or substance abuse issues than their counterparts who sleep in shelters. Thus cross-city comparisons must be made with caution. Likewise, cross-national comparisons should be cautioned due to different economic, health, and social welfare systems. Comparisons of health status of the homeless with the Canadian population and Canadians with incomes less than $15,000 per year were made with available data from the Canadian Community Health Survey (CCHS) Cycle 4.1 (2007) (Statistics Canada 2009).

Results

Demographics

A total of 137 individuals completed the VI survey in October and December 2008, and after eliminating five duplicates, this analysis used data from 132 persons. Table 1 shows that the majority of the participants were male (79.5%) and approximately one-third identified as Aboriginal (First Nations n=30 and Metis n=14). The mean age in the sample was just over 40 years (range: 19 to 72 years old; standard deviation 10.5 years) and the average length of homelessness was just under six years (range: less than 1 month to 30 years). Of the participants, 45.5% (n=60) reported sleeping most frequently in shelters, 19.6% (n=26) reported sleeping most frequently outside, and 18.9% (n=25) reported sleeping in shelters and outside with similar frequency.

Health Conditions

Table 2 presents frequency counts for 12 health conditions and reported substance abuse. Of the 132 participants, excluding substance abuse, 73.0% (n=96) reported having at least one of the health conditions assessed and 53 (55.0%) reported having two or more of the listed health issues. The most frequently reported health conditions were history of frostbite (25.8%), asthma (23.4%), and hepatitis C (22.0%). Of the total participants, 32.6% self-reported current or previous treatment for mental health issues and 96.2% self-reported substance abuse problems.

When the sample is stratified by Aboriginal status, the only significant difference between the two strata is the higher proportion of non-Aboriginal participants who reported having hepatitis C (p=0.032). When the sample is stratified by sex (see Table 2), females had higher frequency counts than males for kidney disease (p=0.0007), liver disease (p=0.0076), heart disease (0.0326), cancer (p=0.0001), hepatitis C (p=0.0176), and mental health issues (p=0.057). No females reported having tuberculosis or HIV.

Comparison with VI Locations and CCHS

Figure 1 shows a comparison of the eight risk indicators for six VI survey locations, including Calgary. Calgary had the highest proportion of reported frostbite, liver disease, and kidney disease compared to the other five locations. Portland and Nashville had a higher proportion of individuals than Calgary who were trimorbid, and reported more than three Emergency Room (ER) visits in the past three months or three ER visits or hospitalizations in the past year.

In a comparison with available data from the CCHS Cycle 4.1 (2007) (Statistics Canada, 2009), the homeless in Calgary had a significantly higher proportion of self-reported health conditions than the general Canadian population for kidney disease (p=0.0001), asthma (p=0.0001), emphysema (p=0.0001), and cancer (p=0.0001). There was no significant difference found with proportion of diabetes between the homeless in Calgary and the general Canadian population (p=0.8901). In comparison to those with an income of less than $15,000 per year (a number chosen by the authors as an indicator of low income) the homeless in Calgary had a significantly higher proportion of self-reported health conditions for kidney disease (p=0.0001), asthma (p=0.0001), emphysema (0.0003), and cancer (p=O.O013) (see Figure 2).

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Discussion

This study provides a description of the health of individuals who were experiencing absolute homelessness and living on the streets of Calgary in October and December 2008. Interpretation of our results should recognize the known caveats of working with data collected in a convenience sample. The sample is not stratified or cross-sectional and thus is not representative of the entire population. It also relied in part on observational reports of the volunteer interviewers as to the presence of substance abuse or mental health issues, which may not be entirely accurate. The reliability and validity is unknown since that sort of assessment was not part of the survey process. Nonetheless, the results do provide valuable information about the health of this hard to reach vulnerable population and can be used to guide and inform interventions and policy initiatives aimed at addressing homelessness.

There are a number of studies demonstrating that homeless individuals carry a heavy burden of disease and have poor health compared to the general population (Hwang 2001; Hwang et al. 2008; Hwang et al. 1998; Frankish, Hwang and Quantz 2005; Kushel, Vittinghoff and Haas 2001). This is confirmed by the results of the VI in Calgary in which 73% of participants (n=96) reported at least one of the listed health conditions and over one-third of individuals reported two or more of the listed health conditions (Table 2). When comparing proportions of self-reported health conditions of the Calgary VI participants to the general Canadian population it is clear that those who are homeless carry a greater burden of disease compared to the general population and even among low-income housed Canadians. Beyond the very high representation of Aboriginal individuals in the homeless population, Aboriginal status is not associated with differential health outcomes compared to non-Aboriginal homeless persons. This finding requires further investigation as it might be a problem with the measure as it does not pick up on unreported or undiagnosed disease; for instance, disproportionately high rates of diabetes and HIV among Aboriginal populations was not picked up with this instrument. This is a limitation in our comparisons also, as unreported or undiagnosed disease might be prevalent in populations of low-income Canadian or Canadians in general.

The Calgary VI survey shows that females are represented in the same relative proportion to the overall homeless population (20.6% compared to 23%) (City of Calgary 2008), which suggests that this sample may be representative of women in the absolute homeless population. Women's health status is much worse than it is for the male homeless in this sample. Females reported a significantly higher proportion of liver disease, kidney disease, heart disease, cancer, hepatitis C, and previous or current mental health treatment compared to their male counterparts, but, as discussed below, they report the same high rate of substance abuse as their male counterparts. A previous cohort study and review of literature reported the age-adjusted mortality rates for young homeless women in Toronto were significantly higher than mortality rates among young women in the general Toronto population (Cheung and Hwang 2004). The present finding of increased rates of health conditions often associated with chronic substance abuse suggests that homeless substance abusing women might be more likely than their male counterparts to experience life-threatening adverse health effects of homelessness. Since this survey was a point-in-time survey of health conditions, it is not possible to determine if these health differences are either, or both, the cause or effect of homelessness (Hwang 2002).

Among the listed health conditions, HIV, history of cold-related injury, renal disease, and liver disease have been associated with a high risk of mortality in homeless populations (Hwang et al. 1998). Calgary had the highest proportion of homeless individuals who reported liver disease, kidney disease, and history of frostbite compared to the other VI locations; suggesting that in addition to carrying a heavy burden of disease, a high number of homeless individuals in Calgary are at an increased risk of dying prematurely. Like the other comparison cities, Calgary appears to have a relatively low proportion of those with HIV/AIDs living on the street. Other studies suggest much higher rates of homelessness for persons with this diagnosis (Walley et al. 2008). This result might be due to the stigma associated with reporting these conditions; and therefore, a possible source of measurement error. Our low rate findings could be an anomaly, specific to Calgary, or a general trend in that the other cities report similar low rates of HIV/ AIDS among this population.

Previous studies also found that male sex, white race, and history of substance abuse increased the risk of death for individuals who are homeless (Hibbs et al. 1994; Hwang et al. 1998). The majority of participants in the Calgary survey were male and 96% reported previous or current substance abuse. This extremely high rate is well in excess of that reported in other studies of Calgary's homeless (CHF 2004) and in other Canadian cities (Goering et al. 2002). Substance abuse is associated with many of the listed health conditions, which may explain the high percentage of asthma, liver, and heart disease in this population. This finding further suggests that a large proportion of this population in Calgary is at an increased risk of death.

Despite the reporting that homeless individuals have poor health, studies have shown it is difficult to identify a causal relationship between homelessness and poor health (Hwang 2002). Individual risk factors and vulnerabilities such as poverty, unsanitary/crowded living conditions, substance abuse, mental health, coping mechanisms, nutritional deficiencies (Whitbeck, Chen and Johnson 2006), and social support networks all contribute to the health and well-being of an individual experiencing homelessness (Frankish, Hwang and Quantz 2005). Structural (societal) factors including local resources, policies, access to services, and interventions are also underlying factors to this association (Jahiel 1992). All of these factors are difficult to measure and evaluate to clearly demonstrate association with poor health (Hwang 2002). However, the comparison of the VI results among the different locations shows that regardless of geographical location and the assumed contextual intricacies within these locations, a high proportion of individuals who are homeless have poor health. This proportion remains high even when compared with low-income, housed Canadians.

Conclusion

Homeless individuals in Calgary appear to carry a heavy burden of disease and as a result might be at an increased risk of dying prematurely. Aside from preventing early mortality, adequate housing could help to relieve health conditions, leading to an improved quality of life, and then potentially reduce overall health care costs. The results of this survey demonstrate that immediate interventions and initiatives are required in Calgary in order to minimize this health, economic, and societal burden and improve the health of these individuals. As a result of the survey in Calgary 34 individuals were identified as vulnerable and at an increased risk of mortality based on the criteria listed above. To date, 26 of those individuals have been housed with the help of the CHF. The VI tool is currently being further developed and utilized in Calgary's 10 year Plan to End Homelessness to direct relevant re-housing interventions (Calgary Committee to End Homelessness 2008).

Although the study was limited by scope and convenience sampling, it provided valuable information about a hard-to-reach population; information that can be used to inform housing interventions and policy initiatives.

References

Aday, L.A. 1994. Health status of vulnerable populations. Annual Review of Public Health 15: 487-509.

Badiaga, S., H. Richet, P. Azas, C. Zandotti, C., F. Rey, R. Charrel, E. Benabdelkader, M. Drancourt, D. Raoult and P. Brouqui. 2009. Contribution of a shelter-based survey for screening respiratory diseases in the homeless. European Journal of Public Health 19 (2): 157-160.

Barrow, S.M., D.B. Herman, P. Cordova and E.L. Struening. 1999. Mortality among homeless shelter residents in New York City. American Journal of Public Health 89 (4): 529-534.

Bebout, R. 1999. Housing solutions: The Community Connections Housing Program. Preventing homelessness by integrating housing and supports. Alcoholism Treatment Quarterly 17 (1/2): 93-112.

Calgary Committee to End Homelessness. 2008. Calgary's 10 year plan to end homelessness, http://www.endinghomelessness.ca/ (Accessed February 28, 2009).

Calgary Homeless Foundation. 2004. Housing our homeless. Calgary, AB: Calgary Homeless Foundation.

Caton, C.L.M. 1990. Homeless in America. New York, NY: Oxford University Press.

Chen, X., L. Thrane, L.B. Whitbeck and K. Johnson. 2006. Mental disorders, comorbidity, and postrunaway arrests among homeless and runaway adolescents. Journal of Research on Adolescence 16 (3): 379-402.

Cheung, A. and S.W. Hwang. 2004. Risk of death among homeless women: A cohort study and review of literature. Canadian Medical Association Journal 170 (8): 1243-1248.

City of Calgary. 2008. The biennial count of homeless persons, http://www.calgary. ca/portal/server.pt/gateway/PTARGS 0 2 428286 0 0 18/Biennial+Count +of+Homeless+Persons.htm (Accessed February 29, 2009).

Common Ground. 2008. Santa Monica service registry creation fact sheet. Santa Monica Government. www01.smgov.net/hsd/pdfs/SM%20Service%20 RegistryFact%20Sheet%20013108.pdf (Accessed February 20, 2009).

Common Ground. 2007. Vulnerability Index. http://www.commonground. org/?page_id=789 (Accessed February 1, 2009).

Common Ground Institute for Park Center Inc and Metropolitan Homelessness Commission. n.d. Vulnerability Index of Nashville's homeless, http://search. tennessean.com/sp?aff= 1102&keywords=trainer&skin (Accessed February 20, 2009).

Common Ground Institute for Portland's Bureau of Housing and Community Development. n.d.. Vulnerability Index fact sheet, www.portlandonline.com/ BHCD/index.cfm?a=217512&c=489 (Accessed October 24, 2008).

D'Amore, J., O. Hung, W. Chiang and L. Goldfrank. 2001.7he epidemiology of the homeless population and its impact on an urban emergency department. Academic Emergency Medicine 8 (11): 1051-1055.

de Vries, G. and R.A.van Hest. 2006. From contact investigation to tuberculosis screening of drug addicts and homeless persons in Rotterdam. European Journal of Public Health 16 (2): 133-136.

Frankish, J.C., S.W. Hwang and D. Quantz. 2005. Homelessness and health in Canada: Research lessons and priorities. Canadian Journal of Public Health 96 (2): S23-S29.

Gelberg, L., R.M. Andersen and B.D. Leake. 2000. The behavioral model for vulnerable populations: Application to medical care use and outcomes for homeless people. Health Services Research 34 (6): 1273-1302.

Gelberg, L., L.S. Linn and S.A. Mayer-Oakes. 1990. Differences in health status between older and younger homeless adults. Journal of the American Geriatric Society 38 (11): 1220-1229.

Goering, P., G. Tolomiczenko, T. Sheldon, K. Boydell and D. Wasylenki. 2002. Characteristics of persons who are homeless for the first time. Psychiatric Services 53 (11): 1472-1474.

Hibbs, J.R., L. Benner, L. Klugman, R. Spencer, I. Macchia, A.K. Mellinger and D.K. Fife. 1994. Mortality in a cohort of homeless adults in Philadelphia. New England Journal of Medicine 331 (5): 304-309.

Hwang, S. 2002. Is homelessness hazardous to your health? Canadian Journal of Public Health 93 (6): 407-410.

Hwang, S. 2001. Homelessness and health. Canadian Medical Association Journal 64 (1): 229-233.

Hwang, S.W. 2000. Mortality among men using homeless shelters in Toronto, Ontario. Journal of the American Medical Association 16 (283): 2152-3257.

Hwang, S.W., A. Colantonio, S. Chiu, G. Tolomiczenko, A. Kiss, L. Cowan, D.A.Redelmeier and W. Levinson. 2008. The effect of traumatic brain injury on the health of homeless people. Canadian Medical Association Journal 179 (8): 779-784.

Hwang, S.W., J.M. Lebow, M.F. Bierer, J.J. O'Connell, E.J. Orav and T. Brennan. 1998. Risk factors for death in homeless adults in Boston. Archives of Internal Medicine 158: 1454-1460.

Kanis, B. 2009. Common Ground. http://www.commonground.org/wpcontent/ uploads/2008/04/Vulnerability%20Index%20101.pdf) (Accessed February 1, 2OO9).

Jahiel, R. 1992. Homeless-making processes and homeless-makers. In Homelessness: a prevention oriented approach, ed. R. Jahiel, 269-296. Baltimore: John Hopkins University Press.

Kessell, E.R., R. Bhatia, J.D. Bamberger and M.B. Kushel. 2006. Public health care utilization in a cohort of homeless applications to a supportive housing program. Journal of Urban Health: Bulletin of the New York Academy of Medicine 83 (5): 860-873.

Kidder, D., R. Wolitski, M. Campsmith and G. Nakamura. 2007. Health status, health care use, medication use, and medication adherence among homeless and housed people living with HIV/AIDS. American Journal of Public Health 97 (12): 2238-2245.

Kim, D.H, C. Daskalakis, J. Plumb, S. Adams, R. Brawer, N. Orr, K. Hawthorne, E.C. Toto and D.J. Whellan. 2008. Modifiable cardiovascular risk factors among individuals in low socioeconomic communities and homeless shelters. Family & Community Health Innovations in Family and Community Health 31 (4): 269-280.

Kim, M., J. Swanson, M. Swartz, D. Bradford, S. Mustillo and E. Elbogen. 2007. Healthcare barriers among severely mentally ill homeless adults: Evidence from the five-site health and risk study. Administration & Policy in Mental Health 34 (4): 363-375.

Kushel, M.B., S. Perry, D. Bangsberg, R. Clark and A.R. Moss. 2002. Emergency department use among the homeless and marginally housed: Results from a community-based study. American Journal of Public Health 92 (5): 778-784.

Kushel, M.B., E. Vittinghoff and J.S. Haas. 2001. Factors associated with the health care utilization of homeless persons. Journal of the American Medical Association 285 (2): 200-206.

Lashley, M. 2007. A targeted testing program for tuberculosis control and prevention among Baltimore city's homeless population. Public Health Nursing 24 (1): 34-39.

LA Health Action. n.d. Project 50. lahealthaction.org/library/Project_50_ briefing.2.13.08.pdf (Accessed February 21, 2009).

Levitt, A.J., D.P. Culhane, J. DeGenova, P. O'Quinn and J. Bainbridge. 2009. Health and social characteristics of homeless adults in Manhattan who were chronically or not chronically unsheltered. Psychiatric Service 60:978-981.

Lewis, J.H.D.O., R.M.P. Andersen and L.M.D.M. Gelberg. 2003. Health care for homeless women: Unmet needs and barriers to care. Journal of General Internal Medicine 18 (11): 921-928.

Martins, D.C. 2008. Experiences of homeless people in the health care delivery system: A descriptive phenomenological study. Public Health Nursing 25 (5): 420-430.

McCormack, D. and J. Macintosh. 2001. Research with homeless people uncovers a model of health. Western Journal of Nursing Research 23 (7): 679-697.

Metsch, L.R., C.B. McCoy, H.V. McCoy, J.M. Shultz, S. Lai, N.L. Weatherby, H. McAnany, R. Correa and R.S. Anwyl. 1995. HIV-related risk behaviors and seropositivity among homeless drug-abusing women in Miami, Florida. Journal of Psychoactive Drugs 27 (4): 435-446.

Morrison, D.S. 2009. Homelessness as an independent risk factor for mortality: Results from a retrospective cohort study. International Journal of Epidemiology 38 (3): 877-883.

Nickasch, B. and S. Marnocha. 2009. Healthcare experiences of the homeless. Journal of the American Academy of Nurse Practitioners 21 (1): 39-46.

Nyamathi, A.M., E.L. Dixon, D. Wiley, A. Christiani and A. Lowe. 2006. Hepatitis C virus infection among homeless men referred from a community clinic. Western Journal of Nursing Research 28 (4): 475-488.

Nyamathi, A.M., B. Leake and L. Gelberg. 2000. Sheltered versus nonsheltered homeless women: Differences in health, behavior, victimization, and utilization. Journal of General Internal Medicine 15: 565-572.

O'Connell, J.J. 2005. Premature mortality in homeless population: A review of literature. Nashville, TN: National Health Care for the Homeless Council, Inc.

Padgett, D., L. Gulcur and S. Tsemberis. 2006. Housing first services for people who are homeless with co-occurring serious mental illness and substance abuse. Research on Social Work Practice 16 (1): 74-83.

Reuler, J.B. 1991. Outreach health services for street youth. Journal of Adolescent Health 12 (7): 561-566.

Ritchey, EJ., M. la Gory and J. Mullis. 1991. Gender differences in health risks and physical symptoms among the homeless. Journal of Health and Social Behavior 32 (2): 33-48.

Savage, C. and C.J. Lindsell. 2008. Health status and access to care for homeless adults with problem alcohol and drug use. Journal of Addictions Nursing 19 (1): 27-33.

Schanzer, B., B. Dominguez, P. Shrout and C.L.M. Caton. 2007. Homelessness, health status, and health care use. American Journal of Public Health 97 (3): 464-469.

Schutt, R.K. and G.R. Garrett. 1988. Social background, residential experience, and health problems of the homeless. Psychosocial Rehabilitation Journal 12: 67-70.

Sherriff, L.C.H. and R.T. Mayon-White. 2003. A survey of hepatitis C prevalence amongst the homeless community of Oxford. Journal of Public Health Medicine 25 (4): 358-361.

Sosin, M.R. and M. Bruni. 1997. Homelessness and vulnerability among adults with and without alcohol problems. Substance Use andMisuse 32 (7/8): 939968.

StataCorp. 2008. Statistical Software: Release 10. College Station, TX: Stata Corporation.

Statistics Canada. 2009. Canadian community health survey Cycle 4.1. Landru. http://www.statcan.gc.ca/cgi-bin/imdblp2SV.pl?Function=getSurvey&SDDS =3226&lang=en&db=imdb&adm=8&dis=2 (Accessed March 21, 2009).

Unity of Greater New Orleans. n.d. National alliance to end hornelessness, www. endhomelessness.org/files/206l_file_Kagel_Targeting_Vulnerability.ppt (Accessed February 10, 2009).

Walley, A.Y., D.M. Cheng, H. Libman, D. Nunes, C.R. Horsburgh, R. Saitz and J.H. Samet. 2008. Recent drug use, homelessness and increased short-term mortality in HIV-infected persons with alcohol problems. AIDS 22 (3): 41542O.

Weinreb, L., R. Goldberg, E. Bassuk and J. Perloff. 1998. Determinants of health and service use patterns in homeless and low-income housed children. Pediatrics 102 (3): 554-562.

Whitbeck, L.B., X. Chen and K.D. Johnson. 2006. Food insecurity among homeless and runaway adolescents. Public Health Nutrition 9 (1): 47-52.

Cherie L. Nicholson

Department of Community Health Sciences

John R. Graham

Faculty of Social Work

J.C. Herbert Emery

Department of Economics and Department of Community Health Sciences

Jeannette Waegemakers Schiff

Faculty of Social Work

University of Calgary

Marina L. Giacomin

Alina I. Tanasescu

Calgary Homeless Foundation

Table 1: Demographics of VI survey participants

Demographic Male n (%) Female n (%)
Characteristic

Sex 105 (79.5) 27 (20.6)
Age 41.3 years 38.5 years
Aboriginal * 35 (33.3) 9 (33.3)
Non-Aboriginal 70 (66.7) 18 (66.67)
Average years homeless 5.8 years 5.7 years
Sleep most frequently **
In shelters 51 (50) 9 (36)
Outside 20 (19.6) 6 (24)
Both shelters and outside 21 (20.6) 4 (16)
Someone else's place 3 (2.94) 1 (4)
Other 7 (9.8) 6 (24)

* First Nations or Metis

** five individuals did not respond to this question

Table 2: Number and Proportion of Individuals with Self-Reported
Health Conditions, and Proportion Non-Aboriginal, Aboriginal,
Male, and Female of Total

Health condition % non-Aboriginal (nA) and
 % (n) Aboriginal (A) [95% CI]

Kidney disease 7.6 (nA) - 8.14 [2.36-13.9]
(including end-stage (10) (A) - 6.98 [0-14.59]
renal disease or dialysis)

Liver disease 13.6 (nA) - 14.0 [6.63-21.28]
(including cirrhosis or (18) (A) - 13.6 [3.5-23.78]
end-stage liver disease)

Heart disease 15.9 (nA) - 16.3 [8.48-24.08]
(including arrhythmia (21) (A) - 15.9 [5.1-26.72]
or irregular heartbeat)

Asthma 23.4 (nA) - 22.1 [13.32-30.86]
 (31) (A) - 27.3 [14.11-40.43]

History of 25.8 (nA) - 29.8 [20.27-39.5]
frostbite (including (24) (A) - 18.2 [6.79-29.58]
hypothermia, foot rot
or immersion foot)

Tuberculosis 2.3 (nA) - 1.2 [0-3.39]
 (3) (A) - 4.6 [0-10.7]

HIV/AIDS 1.5 (nA) - 2.3 [0-5.45]
 (2) (A) - 0.0

Emphysema 5.3 (nA) - 5.8 [0.86-10.64]
 (7) (A) - 4.6 [0-10.7]

Diabetes 4.6 (nA) - 5.8 [0.87-10.76]
 (6) (A) - 2.3 [0-6.68]

Cancer 7.6 (nA) - 8.1 [2.36-13.92]
 (10) (A) - 6.8 [0-14.27]

Hepatitis C 22.0 (nA) - 27.9 [18.43-37-391
 (29) (A) - 11.4 [1.99-20.74]

Substance abuse * 96.2 (nA) - 97.7 [94.61-100]
 (127) (A) - 93.2 [85.73-100]

Mental health ** 32.6 (nA) - 33.3 [23.42-43.24]
 (43) (A) - 31.8 [18.06-5.58]

Health condition Gender (%) - Male (M)
 and Female (F) [95% CI]

Kidney disease (M) - 3.85 [0.151-7.54]
(including end-stage (F) - 24.0 [7.26-40.74]
renal disease or dialysis)

Liver disease (M) - 9.71 [3.99-15.4]
(including cirrhosis or (F) - 29.63 [12.41-46.85]
end-stage liver disease)

Heart disease (M) - 12.62 [6.21-19.03]
(including arrhythmia (F) - 29.63 [12.41-46.85]
or irregular heartbeat)

Asthma (M) - 20.39 [12.61-28.17]
 (F) - 37.01 [18.82-55.25]

History of (M) - 24.04 [15.83-32.25]
frostbite (including (F) -33.33 [15.55-51.11]
hypothermia, foot rot
or immersion foot)

Tuberculosis (M) - 2.9 [0-6.1]
 (F) - 0.0

HIV/AIDS (M) - 1.9 [0-4.56]
 (F) - 0.0

Emphysema (M) - 5.77 [1.28-10.25]
 (F) - 3.7 [0-10.83]

Diabetes (M) - 2.91 [0-6.16]
 (F) - 11.11 [0-22.97]

Cancer (M) - 2.9 [0-6.16]
 (F) - 25.9 [9.4-42.46]

Hepatitis C (M) - 17.48 [10.14-24.81]
 (F) - 40.74 [22.21-59.27]

Substance abuse * (M) - 96.3 [89.17-100]
 (F) - 96.2 [92.53-99.85]

Mental health ** (M) - 28.8 [20.14]
 (F) - 48.1 [29.2-66.99]

* substance abuse refers to those that have ever abused drugs or
alcohol

** mental health refers to those who are currently or have ever
received treatment for mental health issues