Shelter design and service delivery for women who become homeless after age 50.

McLeod, Heath ; Walsh, Christine A.

Abstract

We examine the shelter experiences of women who become homeless for the first time after age 50 using data drawn from interviews with eight women living in shelters in Vancouver, Calgary, Montreal, or Halifax. The findings highlight ways in which shelter site, situation, and service delivery affect older women's shelter experience. Attention to these considerations are important to best meet the needs of this population and help facilitate the transition from homeless to home for women who become homeless in later life.

Keywords; site, situation, service, homeless, women, older adults, shelters

Resume

Nous examinons les experiences de logement des femmes qui deviennent sansabri pour la premiere fois apres 50 ans a l'aide des donnees tirees dentretiens avec huit femmes vivant dans des abris a Vancouver, Calgary, Montreal, ou a Halifax. Les resultats mettent en evidence maniere dont le site de l'abri, de la situation, et la prestation de services affectent lexperience de refuge pour les femmes plus agees. Attention a ces considerations sont importantes pour repondre au mieux aux besoins de cette population et de faciliter la transition de sans-abri a la maison pour les femmes qui deviennent sans-abri dans la vie plus tard.

Mots cles: le batiment, la situation, service, les femmes, sans-abri, les personnes agees, abris

**********

Although income inequality between men and women working in full-time paid employment is decreasing (Drolet 2011), a substantial proportion of women in Canada have low income. Women, particularly women age 45-64, make up a significant proportion of those living in low-income. Over 30% of unattached women age 4554 and 40% of women 55-64 fall under the low-income cut-off (Collin and Jensen 2009). Women are more likely to work part-time jobs (Williams 2010), and out of all Canadians earning minimum wage, 60% are women (Townson 2009). Women were also more likely than men to live in persistent poverty (The Standing Committee on Human Resources, Skills and Social Development and the Status of Persons with Disabilities 2010) and to spend more than 30% of their income on shelter costs (Williams 2010).

Women in Canada experience structural disadvantages across the lifespan including greater exposure to violence and economic marginalization. Women living in poverty are more likely to be victims of physical and sexual assaults (Kushel et al. 2003; Pavao et al. 2007; Wenzel et al. 2004). Histories of abuse, in childhood or as an adult, can lead to homelessness (Callaghan, Farha, and Porter 2002; Crane and Warnes 2001; Hecht and Coyle 2001; Kisor and Kendal-Wilson 2002; Sever 2002, 2009). In some cases, women cannot leave abusive situations because of the lack of other housing options while others live in unsafe or unhealthy accommodations, sacrificing necessities like food, clothing, and medical needs to cover the cost of rent (Callaghan, Farha, and Porter 2002).

Women are more likely to become homeless for the first time in later fife than men (Cohen 1999; McDonald, Dergal, and Cleghorn 2007). As homeless shelters are often the established point of intervention with older homeless adults (Gibeau 2001; McDonald et al. 2006; Warnes and Crane 2000), it is necessary to get a better understanding of the shelter experience and the service requirements of women who become homeless after age 50. The purpose of this paper is to examine how women who become homeless for the first time after age 50 navigate their experience and conceive of the shelter space, and their implications for service delivery in order to provide recommendations designed to better meet the needs of this vulnerable, yet seldom examined population.

Pathways that bring older women to homelessness have been described as crisis-driven (Hecht and Coyle 2001). This hypothesis, however, diminishes the structural elements of older women's homelessness: that individual cases of homelessness are an issue of society and government more than merely a bundle of issues labelled 'personal problems.' Several American studies have examined the relationship between homelessness and victimization. In San Francisco, 30% of women, among the 2,577 adults surveyed, experienced either physical and/or sexual abuse while homeless (Kushel et al. 2003). Wenzel, Koegel, and Gelberg (2000, p. 381), (N = 394), found that mental illness, substance use, and "engaging in economic survival strategies" significantly predicted victimization among the sample of 394 homeless women in Los Angeles. In a longitudinal study comparing homeless women and men (N = 810) in Los Angeles, Wenzel and colleagues (2004) found that violence was a persistent aspect of homeless women's lives, and significantly affected their physical and mental health. Dietz and Wright (2005) examined the occurrence of victimization among older homeless adults (N = 4,200) across the U.S. and found that homeless women experience sexual violence at higher rates than men, and that older women in particular were also at increased risk of physical violence and theft.

Victimization among homeless women has a spatial component. A study examining the geographic occurrences of physical and sexual violence among homeless women in Los Angeles (N = 974) concluded that all women who were homeless shelter consumers experienced physical violence within the shelter's geographic community (Heslin et al. 2007).

In the sole study of older homeless women, Kisor and Kendal-Wilson (2002) analysed adult protective services case records (N = 223) of women age 50 and older who required emergency or temporary shelter in Virginia, USA. They found that older women became homeless through lack of income, poor mental health, interpersonal problems, lack of social supports, and domestic violence. Although the study does not differentiate between those who become homeless in older age, 38% of participants were identified as first-time homeless-service users.

Some research addresses the service needs of older homeless adults, although these do not differentiate by gender, despite the fact that authors argue for the need to consider age and gender simultaneously when designing programs for shelter consumers (Cohen 1999; Chicago Alliance to End Homelessness 2009; Hecht and Coyle 2001). In the US, state-funded programming designated for low-income older adults (age 65 and over) was found to poorly serve the homeless population and programming designed for the general homeless population is inadequate in meeting the unique needs of older homeless adults (Cohen 1999; McDonald et al. 2006). Recommendations for meeting the service needs of older adults emphasize re-training staff to work effectively with the unique needs of the older homeless population (Proehl 2007) and developing coalitions to advocate for greater funding and services so that the population can receive independent and supported housing where appropriate (Gibeau 2001). Kisor and Kendal-Wilson (2002) suggest that older homeless women need emergency shelters with staff to keep women safe from violence, and to provide comprehensive support services designed to foster independent living skills. On the other hand, Lipmann (2009) argues that services for older homeless adults must focus on the 'elderly' rather than 'homeless' aspect of the population, thus changing the focus from sheltering to long-term care.

Adults age 50 through 64, whether newly homeless or not, who seek opportunities to exit homelessness describe issues such as ineligibility for government programs for seniors, and an inability to access subsidized housing or market-based rental housing (Chicago Alliance to End Homelessness 2009; Cohen 1999; Crane et al. 2005; Crane, Warnes, and Fu 2006; Lipman and Rota-Bartelink 2007; McDonald et al. 2006; Shinn et al. 2007).

Evidence suggests that service use depends on whether an older person is newly homeless or not. McDonald and colleagues (2007) compared the experiences of two groups of adults over age 50 (N = 68) in Toronto: individuals who had histories of homelessness and individuals who became homeless for the first time in later life. They found that, besides the preponderance of women, the newly homeless were less likely to be single, and had more children and grandchildren than the long-term homeless group. The first-time homeless group exhibited a lack of knowledge of the homeless-serving system, with more than 70% of the first-time homeless adults unable to find adequate shelter compared with 19% of the long-term homeless adults (McDonald, Dergal, and Cleghorn 2007). Individuals with a longer history of homelessness were far more likely to use emergency medical services (51% compared with 36%) and mobile services (21% and no utilization). The first-time homeless individuals were focused on finding housing, and tended to use housing-related services rather than homeless-sector services, while the longer-term homeless group were less likely to seek housing.

For those becoming homeless in later life, service provision (counselling, outreach, warm clothing, and meals, among others) is often seen as the first step to help mitigate the "culture shock" of homelessness (Gibeau 2001; McDonald et al. 2006; Proehl 2007). As homeless shelters are one of the primary sites housing homeless people and play a major role in the transition between living without shelter and living in stable housing, any improvement made to them will positively affect the clientele (Graham, Walsh, and Sandalack 2008).

Limited research on optimizing shelter design to meet the needs of older homeless persons has been conducted. Some studies have suggested that shelters designed specifically for older adults (of either gender) avoid the noise and overcrowding of all-ages facilities and create a more welcoming atmosphere for older homeless adults (Gibeau 2001; Lipmann 2009; Proehl 2007; Warnes and Crane 2000). Proehl (2007) argues that all facets of agency design should be informed by ideas of social justice. She further suggests that the organizational structure such as a lack of staff hierarchy was appreciated by older homeless adults, who identified that it created opportunities for flexible interactions between staff and client that encouraged active participation in the broader community. Recent studies have begun to investigate the design features of shelters and the subsequent impact on service delivery and community relationships (Graham, Walsh, and Sandalack 2008; Shier, Walsh, and Graham 2007; Walsh et al., 2010). Limited studies gather perspective of service users (Walsh, Rutherford, and Kuzmak, 2009 Walsh et al., 2010).

Further, findings are seldom reported separately for the subgroup that is female, over age 50, and experiencing homelessness for the first time. The available literature focuses on the three main areas: the causes of first-time homelessness over age 50, victimization among the homeless population, and female experiences of poverty and homelessness.

Methods

This study examines the experiences of women entering homelessness after age 50, and their reflections on how shelters for individuals experiencing homelessness can be improved. The sample is drawn from a qualitative, exploratory study regarding optimal homeless shelter design from the perspective of homeless women shelter consumers (Walsh et al. 2010).

Studies use varied definitions of an older adult. Some scholars suggest age 55 (Kutza and Keigher 1991), while others delimit the population based on eligibility for a specific social service, at age 60 (Keigher and Greenblatt 1992) or age 65 (Stergiopoulos and Herrmann 2003). The current thinking among researchers is that age 50 is a useful cut point (Cohen 1999; Crane and Warnes 2001; Crane, Warnes, and Fu 2006; Rota-Bartelink and Lipmann 2007; McDonald, Dergal, and Cleghorn 2007; McDonald et al. 2006; Proehl 2007) since it accounts for the early-aging effects of homelessness (Gelberg, Linn, and Mayer-Oakes 1990) and the early mortality of homeless individuals (Hwang et al. 2009), women in particular (Cheung and Hwang 2004).

The definition of "homeless" also varies among studies. The Canadian Homelessness Research Network (2012) defines homelessness as a situation:

   ... of an individual or family without stable, permanent,
   appropriate housing, or the immediate prospect, means and ability
   of acquiring it. It is the result of systemic or societal barriers,
   a lack of affordable and appropriate housing, the
   individual/household's financial, mental, cognitive, behavioural or
   physical challenges, and/or racism and discrimination, (p. 1)

Importantly, participants' perspectives on what it means to be homeless and their individual experiences differ.

In the larger study (Walsh et al., 2010), a convenience sample (N=76) heterogeneous with respect to ethnicity, sexual orientation, ability, age, and family composition was recruited from Vancouver, Calgary, Toronto, Ottawa, Montreal, and Halifax. Interviews were conducted in English or French; French language interviews were forward and back translated to ensure meaning was preserved. The study received approval from the Conjoint Faculties Research Ethics Board (CFREB) at the University of Calgary. Participants gave written informed consent before providing basic demographic information and participating in individual qualitative face-to-face individual interview about their perceptions of shelters' layout and design, location within a community and city, and the services that they provide, as well as their vision of an 'ideal' shelter. Trustworthiness of the study was established by several means as detailed by Shenton (2004) including the use of well established research methods, triangulation through the use of multiple shelters in six major cities, in Canada, member checking via several group presentations of the preliminary findings, "thick description of the phenomenon under scrutiny" (p. 69), and assessment of congruence with previous research findings. Although the current analysis was not foreseen during the initial interviews, women were made aware that information they provided could be used in a number of different ways to more fully understand their experiences.

The data were analyzed using qualitative methods of analytical induction through a process of coding and thematic analysis (Strauss and Corbin 1998) and the use of Atlas Ti software (Muhr 2004). Researchers read transcripts assigned codes and code descriptors to the codes; they then collapsed, integrated and organized the codes into themes and subsequently identified quotes from the transcripts with pseudonym (or anonymous) and city that illustrated the various themes.

Thematic analysis provided insights into how the interface between site, defined as the shelter's physical design and layout; situation, comprised of the physical location of a building in relation to city zoning and neighbourhoods; and service, the interventions which take place within the shelter, affected older women's experiences of homeless shelters and identified recommendations to better meet their needs within the shelter system.

Eight participants residing in Vancouver, Calgary, Montreal, or Halifax had a self-reported first experience of homelessness after age 50. At the time of the interview, women participants ranged in age from 52 to 63, with an average age of 57. They reported their first experience of homelessness as occurring between 50 and 58 years of age, with an average age of 53. Three participants reported experiences of losing and regaining housing after age 50. Two women identified themselves as Aboriginal; the remainder were Caucasian. Four women were single; others were widowed, separated, divorced, or in common-law relationships. Reported pathways into homeless included the death of a spouse, drug and/or alcohol relapse, family conflict, inability to pay rent, and relocating to a new city.

Findings

Participants provided insights into how the spatial and relational aspects of shelter living affected their experiences of utilizing shelter services for the first time after age 50. The women identified modifications to shelter and program design that would better meet their needs. Three areas require consideration for optimal shelter and program design: shelter site, shelter situation, and service delivery.

Site

Participants identified aspects of shelter design to foster a "normalized" or "comfortable" setting typically found outside of a shelter setting. Common recommendations concerned the improved use of shared spaces, and maintaining a sense of security and privacy. In order to create welcoming shared spaces, respondents suggested improvements such as decor, music, and increased access to laundry facilities.

When shelter sites are recognized by society as worthwhile, some shelter consumers also find it valuable. As one woman noted, "It's a heritage building and to declare it a heritage building I consider that good. It's commendable. I think it puts it in a position that shows the respect for the history of the building" (Monica, Halifax).

Re-interpreting existing shared shelter spaces, such as a "quiet room," was noted as a way of making spaces more welcoming. As Arlene (Halifax) described: "I looked in and thought 'ok, this hasn't been cleaned in weeks, to start with.' There were two hard old brown wooden chairs and a little old table in there. That's not a quiet room, not in my mind." As a way of making the space more welcoming, she continued:

   The space was there but it wasn't utilized. That little quiet room
   that could have been made into a cozy little spot. You could have a
   radio up there, some soft music. Not everyone would use it, I know,
   but some would--A couple of cozy chairs if two girls wanted to
   come--A bunch of artificial flowers ... a couple of books.

For those participants whose histories of shelter use reflected primarily use of violence against women shelters, security was an important part of their experience. Some of the characteristics of shelters that participants identified as contributing to feelings of safety and security included solid doors, peepholes, locked facilities, locked personal storage spaces, and buzzers to enter the site. In addition to these physical characteristics, one participant identified how the design of the space, which required all women entering the shelter to shower together, reduced feelings of safety:

It's an old building, and you feel, I don't know ... all of the women take their showers at the same time. I didn't feel safe taking a shower. You don't have a choice; you had to take a shower when you came in. (Anonymous 1, Montreal)

Security in the shelter was frequently tied to notions of privacy. The ability to secure personal spaces, and having individual or small, shared rooms were reported as means to personally increase women's safety in the shelter and were thought to benefit all shelter consumers. Arlene noted the importance of privacy as it related to personal and collective security: "Everyone [should] have their own private space. Everyone [should] have a place that they knew they could lock any valuables they had and it would never be touched by anyone because that is a big problem." Another woman expanded on the idea, suggesting the importance of privacy as well as feeling at home within the shelter environment:

People [need] to have their own space and their own showers, their own bathrooms. Instead of having to share the bathroom, share the shower and whatever. It would be nice just to have a room with a shower and a hot plate (Joni, Vancouver).

However, not all participants identified privacy within the shelter as desirable. One woman, for example, viewed privacy as a way of creating opportunities for sexual acts that she disapproves of:

It is better the way it is, it's better supervised, if it was in cubicles you wouldn't know what's going on ... there were two women, as a couple, they were making out ... they would be having an orgy by themselves in the room (Anonymous2, Calgary).

Other concerns related to site included adequate heating, increased access to laundry facilities, and the overall cleanliness of the shelter. Most participants discussed the importance of allowing shelter consumers to bring their pets with them into the shelters; two participants had a companion animal at the time of the interview.

Situation

Participants discussed how the shelter situation affected their safety, substance use, and access to basic needs. Respondents were dissatisfied with shelters that were located in close proximity to sources of alcohol and drugs. As Sue (Calgary) articulated:

   When you go to the right out the door there's the liquor store,
   there's the bar, then you go to the left those coconuts are sitting
   over here smoking. So you don't know which way to go so you go
   straight ahead and somebody's offering you a bottle. You can't win
   for losing in that area, I know.

Participants expressed views about shelter location and perceptions of safety. The communities where shelters were located centrally seemed to be run down, with burnt-out street lights, abandoned shops, unsafe parks, limited public transit, payphones out of order, and a lack of police presence. Some respondents heard about, and personally experienced, crimes in the area surrounding the shelter. As a woman (Anonymousl) from Montreal explained:

   Whenever I left [the shelter] I had to be careful. You have drug
   dealers in the park. Across the street from the park there had been
   a murder. In terms of the security outside there, it wasn't much.
   It's rare that I walked around after 8:00 p.m., I preferred staying
   at [the shelter] because it wasn't really safe [outside]. At one
   point I was in a phone booth, someone came inside the phone booth
   with me and he says 'do you have money?' And then I said 'no,' and
   'you get out right now!' You know, the fear, it made me bolder.

Although most participants identified the locations around shelters with negative experiences, respondents also suggested that shelter consumers valued non-judgmental community members. As one woman (Anonymous1) from Montreal stated, "You'd go outside and [community members] wouldn't judge you. There were even people who lived down the street who came to volunteer."

Service

Participants also suggested that service providers had a role to play in establishing relationships with the community. Some suggested that shelter staff must continually engage the community to challenge assumptions about homelessness and living in a homeless shelter. As Monica from Halifax stated, "I think that's important to do hard work with public relations. You have to go forth and say 'well we are doing a good thing here."

Participants in this study identified needs for clothing and laundry services, counselling, knowledgeable referral services, and transitional housing services leading to independent living. Several participants noted that shelters specifically designed to serve older populations would be ideal. As Ms. Dee Eddy from Halifax illustrated:

   The shelter wasn't a good experience for me because it was designed
   for younger people not mature people like me. Lots of children and
   it was very communal living and it was cliquey--if you didn't fit
   in it was uncomfortable.

A prominent area of interest among participants was the relationship between service user and shelter employee. Participants suggested that ideal relationships were built on ideas of equality. Equality was described as staff applying rules evenly to all shelter consumers, and shelter consumers treated as human equivalents to shelter employees. Regarding the inconsistent application of rules in a shelter Arlene from Halifax provided the following example:

   Staff couldn't keep the residents following the rules because the
   upper management put barriers there: 'Well she didn't do [chores]
   today, maybe tomorrow.' Bla bla bla. Well if rules are there to be
   followed, they're to be followed every day.

Women were frustrated by staff who "played favorites" with shelter consumers, giving out preferred chores that allowed some individuals to remain in the shelter during the day, rather than being required to leave the site until the evening.

Another important aspect of equality for participants was the relationship between shelter employees and consumers, as one woman explained:

When you go in there's all these rooms there, all counsellors. And you know what they do? They come in and they join the clients for dinner. Breakfast, dinner and supper. They don't try to separate us. They join the group (Sue, Calgary).

Staff who were non-judgmental, welcoming, and provided opportunities for meaningful participation in shelter life were identified as important. A woman (Anonymousl)

from Montreal linked non-judgment and equality when asked how shelter services should function. "By not judging to start with. [When staff] objectively listen to us, my God ... I think that's what's most important. We sat down as if we were friends and there wasn't anyone who was better or worse than anyone else."

Older women discussed the importance of participation in decision-making and the daily activities of the shelter. Respondents considered staff-facilitated meetings between shelter users and staff, where they could raise views in a safe, respectful atmosphere, to be helpful in resolving interpersonal conflicts. Participants wanted to be involved in planning events, as Madame Water (Montreal) explained, "the energy of [shelter staff] should be to spend half their time taking into account... talking with the people who live there. We have to consult the people for the activities." Shelter consumers wanted to be involved to increase their comfort in a novel situation, as Arlene from Halifax noted:

   I felt lost. I'm in a city that I don't know with no family, no
   friends and I'm living with this bunch of strangers that, ugh. It
   was scary. What saved me was when I went into the kitchen one day
   and asked if I could help. That's what saved me in that place.

Participants also identified the need for a variety of housing options for women who find themselves homeless in later fife without children, social supports, employment, or incomes sufficient to maintain stable housing.

Discussion

The experiences of women who became homeless in later life were diverse. This study identified causes of first-time homelessness among women over 50 that bore similarity to other studies examining causes of homelessness among older populations. Widowhood, family breakdowns, substance use, rapid deterioration of health, interpersonal conflict, and financial issues were among the causes of homelessness for older adults in other studies (Crane et al. 2005; Kisor and Kendal-Wilson 2002; McDonald, Dergal, and Cleghorn 2007). Widowhood and interpersonal conflicts contributed to the participants' lack of income. Considering that more women become homeless in later life (Cohen 1999; Crane and Warnes 2001; McDonald, Dergal, and Cleghorn 2007), these findings, although based on a small sample, emphasize the importance of preventive measures to keep older women housed.

Similar to Kisor and Kendal-Wilson's (2002) study, participants in this study experienced poverty in both social and economic aspects of life--in some cases, estranged from family members with few individuals to provide informal social support aside from agency staff.

Participants made many recommendations about specific facets of shelter life related to site, situation or service. Similar to findings from other research on the housing needs of homeless women (Labrecque and Walsh 2011; Walsh et al. 2010; Walsh, Rutherford, and Kuzmak, 2009), women who became homeless in later life wanted their shelters to feel less institutional and described ways of creating spaces more consistent with a 'home' than a temporary residence. These included desires for welcoming, comfortable spaces, privacy, security and the provision for companion animals. The need for privacy was emphasized as a boundary with younger shelter consumers whom participants wanted distance from. As older women are the most likely to be victims of theft or physical assault among the homeless population (Dietz and Wright 2005), the concern expressed by women for privacy and security is particularly salient.

The location of shelters was less straightforward. The concept of safety was frequently cited both in terms of personal threat and easy access to drugs and alcohol that compromised participants' safety. However, participants tended to view the surrounding neighbourhoods--all urban, downtown shelters--favourably because they felt that the shelters fit into the surrounding community and faciltated access to other services which were located in core urban areas. Participants described the importance of non-judgment by individuals in the surrounding community and the role that shelter employees had to play in encouraging local acceptance of the shelter and its consumers. These views reflect findings in the literature, where there does not seem to be a clear relationship between community setting and shelter consumer's perceptions. For example, in Los Angeles, shelter consumers were more likely to be victimized close to their shelters, but they were also aware of "safe havens" and problematic streets to avoid (Heslin et al. 2007).

Older women expressed uncertainty about their new role as a homeless shelter consumer, describing it as a "cultural shock", a new experience to which they had difficulty adjusting. As the participants were, in many cases, new to the shelter system, their observations and experiences reflect personal histories that include decades of receiving an income and being housed. Similarly, McDonald and colleagues (2007) discussed the culture shock of first-time homeless adults, particularly the trauma of witnessing crimes. The participants in this study shared similar concerns about witnessing and experiencing crimes in shelters and neighbouring communities.

Regarding services, relationships between shelter staff and consumers were the dominant topic for women in this study. Like those in the larger study, participants identified respect as a fundamental need in the relationship between staff and client. Previous research has suggested that service provision for homeless older adults requires building respect (Canada Mortgage and Housing Corporation 2007; Daiski 2007; Hecht and Coyle 2001; McDonald et al. 2006). Participants in the study also valued autonomy and expressed views that suggested that their ability to make decisions for themselves was compromised as a consequence of their becoming homeless. The staff-consumer relationship has three, somewhat overlapping, components: equality, non-judgment, and participation. Women who discussed equality usually framed it negatively--as a way of ensuring that no shelter consumers receive occasional beneficial treatment rather than encouraging shelter staff to give equal benefit to all shelter consumers. This "negative equality" was also described as using shelter staff to react to situations with harsh punishment for nonconformers. Non-judgment among shelter staff and consumers was important as it emphasized the more common understanding of equality, where women framed it as seeing everyone as an equal human being where all people are deserving of kindness, respect, and honesty. Meaningful ways of participating in shelter life are more possible in a non-judgemental context: from more passive activities of giving feedback to staff via surveys to more active consultation where shelter consumers plan their own events. When women discussed their experiences of participating in shelter operations, from helping in the kitchen to times when shelter staff acted on their recommendations, they identified feelings of acceptance, equality, and respect. The nature of what constitutes 'meaningful' participation needs further research.

Unlike other studies (Cooper, Walsh, and Smith 2009) and the larger study (Walsh et ah, 2010), participants in this study did not report a need for gender-segregated shelters. This may be an artefact of the small sample size or the relationship status of the women, only one of whom was in a relationship. Further, half of the participants had only ever experienced violence-against-women shelters, which house only women or women and children. Thus, they may not have considered males to be an issue in shelters simply because it had never been a part of their experience. There was a marked dissatisfaction for residing in a shelter with younger women with children or women attempting to reunite with their children. This finding is consistent with British and Australian research that found older homeless adults prefer to stay in shelters with individuals of similar age (Crane and Warnes 2001; Lipmann, Mirabelli, and Rota-Bartelink 2004).

This study was limited by a small sample size. As well, half of the women had stayed in a domestic violence shelter; which in its broadest sense could be defined as an earlier episode of homelessness. It is possible, too, that some women may have been inadequately housed in the past; however, the women themselves reported that their first episode of homelessness occurred after age 50. Studies designed to understand the trajectories for women experiencing 'homelessness' for the first time in old age should gather narratives over the life course to more readily account for their histories of housing instability and provided more detailed accounts for their pathways into homelessness in order to inform effective prevention and intervention strategies.

Also, as the main focus of the research was on shelter design, it was not possible to explore aspects of aging and homelessness specifically. In particular, data regarding physical health, accessibility, and access to and use of income supports would have contributed to a more fully developed understanding of older women's experience.

Conclusion

As a consequence of Canada's gendered wage gap, women continue to experience higher rates of low income and the vulnerability associated with living in poverty, including an increased risk of becoming homeless in later life. Thus there is a greater need for understanding older women's experience of homelessness and the various interventions designed to remediate homelessness among this population including homeless shelters. The participants in this study have described women's experiences of ideal shelter site, situation, and service delivery and provide insight on how design features affect feelings of home, privacy, safety, acceptance, equality, and the respectful relationships participants desired. Addressing these recommendations could lead to increased service utilization that would facilitate older women's transition from shelter to home.

References

Bridgman, R. 2006. StreetCities: Rehousing the homeless. Toronto: Broadview Press.

Callaghan, M., L. Farha and B. Porter. 2002. Women and housing in Canada: Barriers to equality. Women's Housing Program, Centre for Equality Rights in Accommodation. Retrieved on Nov. 15,2010 from: http://www.equalityrights. org/cera/docs/barriers.htm

Canada Mortgage and Housing Corporation. 2007. Supportive housing for homeless and hard-to-house seniors: An in-depth case study. (Socio-economic series 07-017). Ottawa: Author.

Canadian Homelessness Research Network. 2012. Canadian definition of homelessness. Homeless Hub. Retrieved from www.homelesshub.ca/CHRNhomelessdefinition/

Chicago Alliance to End Homelessness. 2009. Homeless over 50: The graying of Chicago's homeless population. Chicago: Chicago Alliance to End Homelessness. Cohen, C. 1999. Aging and homelessness. The Gerontologist 39(1): 5-14.

Collin, C and Jensen, H. 2009. A statistical profile of poverty in Canada. Ottawa: Library of Parliament.

Cooper, J., C. A. Walsh and P. Smith. 2009. A part of the community: Conceptualizing shelter design for young pregnant homeless women .Journal of the Association for Research on Mothering 11(2): 122-133.

Crane, M., K. Byrne, R. Fu, B. Lipmann, F. Mirabelli, A. Rota-Bartelink, A. Warnes. 2005. The causes of homelessness in later life: Findings from a 3-nation study. Journal of Gerontology 60B(3): 152-159.

Crane, M., and A. Warnes. 2001. Older people and homelessness: Prevalence and causes. Topics in Geriatric Rehabilitation 16(4): 1-14.

Crane, M., A. Warne and R. Fu. 2006. Developing homelessness prevention practice: combining research evidence and professional knowledge. Health and Social Care in the Community 14(2): 156-166.

Cresswell, T. 1997. In place/out of place: Geography, ideology and transgression. Minneapolis: University of Minnesota Press.

Daiski, I. 2007. Perspectives of homeless people on their health and health needs priorities. Journal of Advanced Nursing 58(3): 273-281.

Dietz, T., and J. Wright. 2005. Age and gender differences and predictors of victimization of the older homeless. Journal of Elder Abuse and Neglect 17(1): 37-60.

Drolet, M. 2011. Why has the gender wage gap narrowed? Ottawa: Statistics Canada.

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 Geriatrics Society 38(11): 1220-1229.

Gibeau, J. 2001. Home free: An evolving journey in eradicating elder homelessness. Topics in Geriatric Rehabilitation 17(1): 22-52.

Graham ,J.R.,C. A. Walsh and B.Sandalack.2008. Homeless shelter design: Considerations for shaping shelters and the public realm. Calgary: Detselig Enterprises.

Hecht, L., and B. Coyle. 2001. Elderly homeless: A comparison of older and younger adult emergency shelter seekers in Bakersfield, California. American Behavioral Scientist 45(1): 66-79.

Heslin, K., P. Robinson, R. Baker, and L. Gelberg. 2007. Community characteristics and violence against homeless women in Los Angeles County. Journal of Health Care for the Poor and Underserved 18(1): 203-218.

Hwang, S., R. Wilkins, M. Tjepkema, P. O'Campo, and J. Dunn. 2009. Mortality among residents of shelters, rooming houses, and hotels in Canada: 11 year follow-up study. British Medical Journal 339: 1-9.

Keigher, S., and S. Greenblatt. 1992. Housing emergencies and the etiology of homelessness among the urban elderly. The Gerontologist 32(4): 457-465.

Kisor, A., and L. Kendal-Wilson. 2002. Older homeless women: Reframing the stereotype of the bag lady. Affila 17(3): 354-370.

Kushel, M. B., J. L. Evans, S. Perry, M.J. Robertson, and A. R. Moss. 2003. No door to lock: Victimization among homeless and marginally housed persons .Archives of Internal Medicine 163(20): 2492-2499.

Kutza, E., and S. Keigher. 1991. The elderly 'new homeless': An emerging population at risk. Social Work 36(4): 288-293.

Laird, G. 2007. Homelessness in a growth economy: Canadas 21st century paradox. Calgary: Sheldon Chumir Foundation for Ethics in Leadership.

Labrecque, J. and C. A. Walsh. 2011. Homeless womens voices on incorporating companion animals into shelter services. Anthrozoos: A Multidisciplinary Journal of The Interactions of People & Animals 24(1): 79-95.

Lipmann, B. 2009. Elderly homeless men and women: Aged care's forgotten people. Australian Social Work 62(2): 272-286.

Lipmann, B., F. Mirabelli and A. Rota-Bartelink. 2004. Homelessness among older people: A comparative study in three countries of prevention and alleviation. Melbourne: Wintringham.

McDonald, L., J. Dergal and L. Cleghorn. 2007. Living on the margins: Older homeless adults in Toronto.,Journal of Gerontological Social Work 49(1/2): 19-46.

McDonald, L., P. Donahue, J. Janes and L. Cleghorn. 2006. In from the streets: The health and well being of formerly homeless older adults. Toronto: Institute for Life Course and Aging.

McInturff, K. 2013. Closing Canada's gender gap. Toronto: Canadian Centre for Policy Alternatives.

Muhr, T. 2004. User's manual for ATLAS. Ti 5.0. (2nd ed.). Berlin, Germany: Scientific Software Development.

O'Reilly-Fleming, T. 1993 .Down and out In Canada: Homeless Canadians. Toronto: Canadian Scholars' Press.

Pavao, J., J. Alvarez, N. Baumrind, M. Induni and R. Kimerling. 2007. Intimate partner violence and housing instability. American Journal of Preventative Medicine 32(2): 143-146.

Proehl, R. 2007. Social justice, respect, and meaning-making: Keys to working with the homeless elderly population. Health and Social Work 32(4): 301-307.

Rota-Bartelink, A., and B. Lipmann. 2007. Causes of homelessness among older people in Melbourne, Australia. Australian and New Zealand Journal of Public Health 31(3): 252-258.

Shenton, A. K. 2004. Strategies for ensuring trustworthiness in qualitative research projects. Education for Information 22: 63-75.

Sev'er, A. 2002. A feminist analysis of flight of abused women, plight of Canadian shelters: Another road to homelessness. Journal of Social Distress and the Homeless 11(4): 307-324.

Sever, A. 2009. More than wife abuse that has gone old: A conceptual model for violence against the aged in Canada and the US .Journal of Comparative Family Studies 40: 279-292.

Shier, M., C. Walsh and J. Graham. 2007. Conceptualizing optimum homeless shelter service delivery: The interconnection between programming, community, and the built environment. Canadian Journal of Urban Research 16(1): 58-85.

Sibley, D. 1995. Geographies of exclusion. New York: Routledge.

Shinn, M., J. Gottlieb, J. Wett, A. Behl, A. Cohen, and D. Ellis. 2007. Predictors of homelessness among older adults in New York City: Disability, economic, human and social capital and stressful events .Journal of Health Psychology 12(5): 696-708.

Snow, D. 2008. A roof over our heads: Affordable housing and homelessness policy in Canada. Calgary: Canada West Foundation.

Standing Committee on Human Resources, Skills and Social Development and the Status of Persons with Disabilities. 2010. Federal poverty reduction plan: Working in partnership towards reducing poverty in Canada. Ottawa: House of Commons.

Statistics Canada. 2006. Women in Canada: A gender-based statistical report. (Catalogue no. 89-503-XIE). Ottawa: Author.

Statistics Canada. 2010. Population projections for Canada, provinces and territories: 2009 to 2036. (Catalogue no. 91-520-X). Ottawa: Author.

Stergiopoulos, V., and N. Herrmann. 2003. Old and homeless: A review and survey of older adults who use shelters in an urban setting. Canadian Journal of Psychiatry 48(6): 374-380.

Strauss, A., and J. Corbin. 1998. Basics of qualitative research. Thousand Oaks, CA: Sage Publications.

Townson, M. 2009. Canadian women on their own are poorest of the poor. Toronto: Canadian Centre for Policy Alternatives.

Walsh, C.A., K. Beamer, C. Alexander, M.L. Shier, M. Loates and J.R. Graham. 2010. Listening to the silenced: Informing homeless shelter design for women through investigation of site, situation, and service. Social Development Issues 32(3): 35-49.

Walsh, C.A., J. R. Graham and M. Shier, M. 2009. Towards a common goal for shelter services: Defining 'best practices.' Social Development Issues 31(2): 57-69.

Walsh, C.A., G.E. Rutherford and N. Kuzmak. 2009. Characteristics of home: Perspectives of women who are homeless. The Qualitative Report 14(2): 299-317.

Warnes, A., and M. Crane. 2000. Achievements of a multiservice project for older homeless people. The Gerontologist 40(5): 618-626.

Wenzel, S., P. Koegel and L. Gelberg. 2000. Antecedents of physical and sexual victimization among homeless women: A comparison to homeless men. American Journal of Community Psychology 28(3): 367-390.

Wenzel, S., J. Tucker, M. Elliott, G. Marshall and S. Williamson. 2004. Physical violence against impoverished women: A longitudinal analysis of risk and protective factors. Women's Health Issues 14(5): 144-154.

Williams, C. 2010. Economic well-being. Ottawa: Statistics Canada.

Heath McLeod and Christine A. Walsh

Faculty of Social Work

University of Calgary