Version classiqueVersion mobile
OpenEdition Books

Homelessness & Health in Canada

 | 
Manal Guirguis-Younger
, 
Ryan McNeil
, 
Stephen W. Hwang

Part I-Homelessness & Health in Canadian Populations

Chapter 5. Homeless Immigrants’ and Refugees’ Health over Time

Fran Klodawsky, Tim Aubry et Rebecca Nemiroff

Texte intégral

Introduction

1Gaps in health research in Canada have been noted with regard to both persons who have experienced homelessness (Frankish, Hwang and Quantz 2005) and immigrants and refugees (Beiser 2005; Dunn and Dyck 2000). Recently, some investigations have reported research on immigrants and refugees who have been homeless (Chui et al. 2009; Kappel Ramji 2002; Klodawsky et al. 2005, 2007; Paradis et al. 2008). In particular, Chui and colleagues' (2009) Toronto-based research on the health of homeless immigrants is the first peer-reviewed published research on this topic in Canada. One important question investigated in that article is whether the healthy immigrant effect, as noted more generally among Canada's newcomer populations, is relevant to homeless immigrants. The healthy immigrant effect is a tendency among newcomers to arrive in Canada (or the United States) healthier than their native-born peers but then for their health status to decline in proportion to the time spent as an immigrant (Ali, McDermott and Gravel 2004; Argeseanu Cunningham, Ruben and Narayan 2008; Beiser 2005; Dey and Lucas 2006; Fennelly 2007; Huh, Prause and Dooley 2008; McDonald and Kennedy 2004; Newbold 2005; Singh and Hiatt 2006). Reasons for the decline remain unclear. Explanations range from adopting lifestyles that are less healthy than those practiced in the country of origin to the adverse impact of stress related to discrimination, poverty and/or unexpected difficulties in finding suitable employment (Simich et al. 2005; Singh and Hiatt 2006). Some researchers also highlight the need to further explore the ways that the heterogeneity of immigrant backgrounds and characteristics intersects with health status, health service utilization and changes over time (Ali, McDermott and Gravel 2004; Argeseanu Cunningham, Ruben and Narayan 2008; Beiser 2005; Castro 2008; Dunn and Dyck 2000).

2Chui and colleagues' analysis confirms that the healthy immigrant effect is also relevant to Toronto's homeless population, with recent immigrants being "physically and mentally healthier and less likely to have chronic conditions and substance use problems than native-born homeless individuals" (2009: 946). They note the significance of length of time since entry to Canada: "the health status of homeless individuals who immigrated more than 10 years ago is not significantly different from that of homeless non-immigrants" (946). That study is significant in that it was based on a large, representative sample of English-speaking homeless immigrants and native-born homeless persons in Toronto. Limitations include its cross-sectional nature and its inability to include refugees and immigrants who were not able to complete the survey in English.

3The primary goal of this chapter is to contribute further insights to that of Chui and colleagues (2009) as well as other Canadian health research about immigrants and refugees that have experienced homelessness. Here, we report on an analysis of responses by both foreign-born and Canadian-born participants who were homeless in 2002 or 2003 and who were participants in a longitudinal study in Ottawa, Ontario: the Panel Study on Homelessness in Ottawa. The Panel Study's objective was to examine the diversity of pathways that lead people into and out of homelessness over time by following a mixed group of individuals and adults with children in Ottawa who were homeless at the beginning of the study. Additional goals were to explain factors that distinguished those who successfully exited homelessness from those who remained homeless or experienced multiple episodes of homelessness and to assess the impacts of their pathways on health (Aubry et al. 2003, 2007; Klodawsky et al. 2007). The Panel Study offers a unique opportunity to contribute knowledge about changes over time in homeless immigrants' and refugees' health status and about their use of health services in a mid-sized Canadian city. It is one of only a few Canadian longitudinal studies having to do with people who have been homeless, and it is the only extant study that offers the opportunity to examine what happens to homeless immigrants and refugees over time.

Methods

Setting and Study Population

4Ottawa is Canada's national capital and fourth largest city, with about 1 million people located in the census metropolitan area (which includes Gatineau, Quebec). The City of Ottawa has become an attractive alternative to Toronto, Montreal and Vancouver for many newcomers. Between 1996 and 2001, Ottawa's immigrant population grew at almost twice the rate of its Canadian-born population. The 2009 report of the Community Foundation of Ottawa (cfo) observed that "foreign-born persons accounted for 22.3 percent of Ottawa's population, or 178,545 persons in 2006 compared to 21.8 percent in 2001" (cfo 2009: 3). The report also highlighted a growing gap between rich and poor, the high cost of housing and health deficiencies as areas of concern that required particular attention (cfo 2009).

5The Panel Study researchers defined being homeless in a manner consistent with other North American academic literature (Susser, Moore and Link 1993) and the City of Ottawa's municipal government: "a situation in which an individual or family has no housing at all, or is staying in a temporary form of shelter" (Region of Ottawa-Carleton 1999: 2). Based on research indicating that the overwhelming majority of homeless adults in Ottawa used emergency shelters at some point (Farrell, Aubry and Reissing 2002), we decided to recruit adult participants from the emergency shelters. On the other hand, because use of emergency shelters by homeless youth was more variable (Farrell, Aubry and Reissing 2002), we specified that half the sample would be drawn from among those who were using homeless services, such as drop-ins, but were not staying in an emergency shelter.

6The Panel Study's interest was in capturing a diversity of experiences of individuals who had been homeless rather than in capturing a representative sample of Ottawa's homeless population overall. As a result, we sought a representative sample from within each of five equal-sized groups of individuals who were homeless at the time of the first set of interviews: adult men, adult women, male youth, female youth and adults accompanied by at least one child under 16. Within each group, the sampling frame was based on available emergency shelter data. Sampling criteria were established as follows: length of stay, in the case of adult women and men; and citizenship status, in the case of adult women and adults with at least one child under 16. After assessing the available data, we specified that 40 percent of adults with children and 25 percent of adult women for the study would not be Canadian citizens and that cultural interpretation would be offered to reduce the likelihood of language as a barrier to being interviewed. Selective sampling was not required in the case of youth: we were able to interview all male and female youth who were eligible and who accepted an invitation to participate (see Table 5-1).

7Four hundred and twelve face-to-face interviews took place between October 2002 and June 2003, and 255 individuals were interviewed again between October 2004 and June 2005. At baseline, 99 respondents were foreign-born, and 58 of them participated in follow-up interviews. The baseline interviews were carried out by 11 trained interviewers with a background in clinical psychology and/or experience in working with homeless people. These interviewers conducted 356 interviews in English, 30 in French and 14 in Somali. The services of cultural interpreters were used for 16 additional interviews (Somali, 4; Arabic, 5; Spanish, 3; Cantonese, 1; Lingala, 1; Russian, 1; and Ukrainian, 1). At follow-up, there were 221 interviews in English, 13 in French and 13 in Somali. The services of cultural interpreters were used for eight other interviews (Arabic, 4; Spanish, 2; Somali, 1; and Cantonese, 1). Their countries of origin are summarized in Table 5-2.

Table 5-1. Overview of panel study subgroups, sampling criteria, number of recruitment sites and baseline and follow-up sample size

Table 5-1. Overview of panel study subgroups, sampling criteria, number of recruitment sites and baseline and follow-up sample size

*Youth were recruited at shelters and drop-in facilities in equal numbers

Interview Protocol and Content

8Managers at the emergency shelters and drop-in centres that we visited and consulted with were very enthusiastic about the study. Their support was vital to the study's success since respondents were recruited with the help of staff at these facilities, who were the first point of contact. Interviewers followed up staff referrals and explained to potential respondents the purpose of the study and its informed consent provisions (as approved by the University of Ottawa Research Ethics Board for the Humanities and Social Sciences). Shelter staff also facilitated the availability of private interview spaces at the shelters. At follow-up, respondents were contacted directly by specially trained 'trackers'. In some cases, these trackers arranged interviews in advance but in many instances, the trackers conducted the interviews themselves as quickly as possible after contact had been made with an interested respondent.

9These interviews took place in private spaces in a variety of locations, including community centres, emergency shelters and offices at the Centre for Research on Community Services at the University of Ottawa. The trackers contacted respondents on the basis of information gleaned from a variety of means that had been approved by respondents at baseline, including address and telephone contact information obtained through the City of Ottawa's Ontario Works files or from family and friends (Aubry et al. 2004). Typically, each interview took between 50 and 150 minutes, with an average length of about 75 minutes. At follow-up, 36 follow-up interviews took place over the phone with individuals who no longer lived in Ottawa. These interviews tended to be about 20 minutes longer than the face-to-face interviews. Individuals were paid an honorarium of $10 for the first interview and $20 for the second interview.

10The interview instruments consisted of both quantitative and qualitative measures. To assess respondents' self-reported health status, we relied on a widely used instrument called the sf-36. This instrument provided a measure of physical health and mental health relative to the US general population, matched on the basis of age and sex (Ware, Kosinski and Gandek 2002). We also asked a series of questions about chronic conditions and injuries that were part of the National Population Health Survey, a longitudinal survey of over 17,000 households across Canada about the current state of health, contact with health-related service providers and health care needs. To this we added questions about other physical health and mental health chronic conditions that were likely relevant to a population that had experienced homelessness. In order to assess the extent and severity of alcohol and drug use problems, we relied on cage, a four-item scale identifying the presence of alcohol use problems (Chan, Pristach and Welte 1994; Mayfield, McLeod and Hall 1974) and the Drug Abuse Screening Test (dast), a 20-item scale identifying the presence of drug use problems (Skinner 1982). Qualitative measures were created and integrated into the interview protocol in order to provide more in-depth information and to provide participants with an opportunity to share their experiences and perceptions.

Table 5-2. Country of origin of foreign-born respondents

Country of Origin

Foreign-Born
Respondents
(
n = 98)

Immigrants
(
n = 45)

Refugees
(
n = 53)

Somalia

22

6

16

United States

8

8

Haiti

6

5

1

Unknown/Missing

6

4

2

Rwanda

5

5

Djibouti

4

4

Zaire

3

2

1

Ethiopia

3

1

2

Columbia

3

3

Congo, Dem. Rep. of

3

3

Burundi

2

2

Italy

2

1

1

Kenya

2

2

Lebanon

2

1

1

Palestine

2

2

Philippines

2

1

1

Ukraine

2

2

Africa (unspecified)

1

1

Angola

1

1

Armenia

1

1

Burkina Faso

1

1

China

1

1

Congo, Republic of

1

1

England

1

1

Eritrea

1

1

Guatemala

1

1

India

1

1

Kuwait

1

1

Poland

1

1

Saudi Arabia

1

1

Scotland

1

1

Singapore

1

1

South Korea

1

1

Sudan

1

1

The Gambia

1

1

Trinidad

1

1

Vietnam

1

1

Yemen

1

1

11In the baseline interviews, we asked respondents whether they were Canadian citizens and whether they had been born in Canada. For those born elsewhere, we also asked when they had arrived in Canada and in Ottawa, and why they moved to Canada. During the course of an investigation of the baseline results in which we compared Canadian-born and foreign-born respondents, we realized that an important question had been missed, having to do with whether respondents had arrived as immigrants or as refugees (Klodawsky et al. 2005). To compensate for this limitation, we revisited the interviews and were able to categorize foreign-born respondents as immigrants or refugees on the basis of their qualitative responses to such questions as why they had come to Canada. In other words, we assessed their qualitative responses overall to answer the question, "does the evidence suggest that this individual came to Canada as an immigrant or a refugee, or is there insufficient evidence to make a decision either way?"

Statistical Analyses

12Our analysis of the baseline interviews revealed that the socio-demographic characteristics of refugees and immigrants in the study were distinct from their Canadian-born peers. Whereas the former were more likely to be female and living with children, Canadian-born respondents were more likely to be men on their own. The reasons for being homeless also differed, with the former more likely to be homeless for economic reasons. To address and compensate for these differences, comparisons were made on the basis of matched pairs

13The matches were selected using the following criteria: (1) participation in both the baseline and follow-up interview and (2) paired matching based on age (adult or youth), sex (male or female) and family status (with or without at least one child under 16). In matching by pairs, random selection among Canadian-born participants was used within subgroups that were made up of these matching variables.

Results

14The results reported here are based on interviews with 90 respondents—45 Canadian and 45 foreign-born—who were matched according to the criteria and approach described above and who completed the follow-up interviews. These matched respondents included: 1 pair of adult men, 11 pairs of adult women, 6 pairs of female youth, 4 pairs of male youth, 12 pairs of women in families and 4 pairs of men in families.

15The quantitative, health-related results of the Canadian-born and the foreign-born matched samples are summarized in Table 5-2. A series of repeated measure anovas were conducted to examine for differences between the two groups, for changes across time and for interactions of group and time on health-related variables measured with continuous data. A series of chi-square analyses were conducted on data presenting the percentage of alcohol and drug use among participants of the two groups at the two time points.

  • 1 Standardized scores involve converting the raw score to a scaled score based on a normative sample (...)

16Overall, foreign-born respondents reported a significantly higher level of mental health functioning than their Canadian-born peers. At follow-up, both Canadian and foreign-born respondents showed a significant and similar level of improvement in mental health functioning However, the average score of both groups of respondents on the sf-36 showed them to have poorer mental health functioning when compared with the average score of the normative sample drawn from the US general population.1

17With regard to physical health, foreign-born participants reported a significantly higher level of physical health functioning than Canadian-born participants across the two time points. There were no changes in level of physical functioning over time with both groups showing similarly stable levels at both time points. In comparison to the US normative group, foreign-born participants demonstrated slightly better physical health than the normative American sample. In contrast, Canadian-born participants reported a similar level of physical functioning to the American sample.

18In line with the differences in the level of physical health functioning, Canadian-born participants also reported having significantly more chronic health conditions at baseline and at follow-up than foreign-born participants. No changes in the number of chronic health conditions were evident over time for either of the groups. Over the two data collection points, foreign-born participants had significantly less contact with service providers in the previous 12 months than Canadian-born participants. Although the utilization of health care providers increased for foreign-born participants, the interaction between groups and time was not significant.

19A significantly higher proportion of Canadian-born participants reported drug use problems at baseline than foreign-born respondents. A larger proportion of foreign-born participants (26%) reported not having a health card compared to Canadian-born participants (13%) at the baseline interview, although this difference was not statistically significant. On the other hand, a higher proportion of Canadian-born respondents (20%) indicated at baseline, when they were homeless, that they had had an experience of not receiving health care when they needed it, compared to foreign-born participants (15%). Again, these differences were not statistically significant.

20The quantitative, health-related results of the immigrants and refugees making up the foreign-born participant group are presented in Table 5-3. A series of repeated measure anovas were also conducted to examine for differences between the two groups, for changes across time and for interactions of group and time on health-related variables measured with continuous data. As well, a series of chi-square analyses were conducted on data presenting the percentage of alcohol and drug use among participants of the two groups at the two time points.

Table 5-3. Health-related quantitative results for panel study matched Canadian-born and foreign-born respondents, and immigrant and refugee respondents

Table 5-3. Health-related quantitative results for panel study matched Canadian-born and foreign-born respondents, and immigrant and refugee respondents

aF(Group) (1, 88) = 6.79, p < 0.02; bF(Time) (1, 88) = 8.06, p < 0.01; cF(Group) (1, 88) = 9.01, p < 0.005;
dF(Group) (1, 90) = 22.01, p < 0.001; eF(Group) (1, 87) = 7.41, p < 0.01; fChi-square (df = 1) = 4.09, p < 0.05

21Our analyses found no differences across the two time points between immigrants and refugees in terms of reported levels of physical health functioning or mental health functioning, number of chronic health conditions and number of health care providers seen in the past year. There was a significant increase in the number of health care providers over time for the two groups together. There were no differences between immigrants and refugees in terms of the reported prevalence of alcohol use or drug use problems at baseline or at the two-year follow-up. Changes over time within the group were also non-significant statistically.

22We also examined the health related, qualitative results of the interviews with adults in families (12 pairs) and single adult women (11 pairs). The qualitative analysis in this study was conducted with the help of atlas.ti, a software package for the analysis of qualitative data (http://www.atlasti.com/​). atlas.ti is organized around the capacity to assign codes to words or phases in the qualitative data that are relevant to the research questions under discussion. In this study, 33 sets of questions at baseline and six at follow-up included some qualitative elements (Aubry et al. 2007). We began by examining systematically the respondents' answers to open-ended questions that directly addressed health care to assess the extent of differences and similarities in the responses of those born in Canada and those born elsewhere. Key questions included, "Have problems with your physical health contributed to your being/becoming homeless? If yes, what are they and how have they contributed?" and "Have you ever been told by a health care professional that you had mental health problems? If yes, how were they explained to you?" In the course of this examination, it became clear that a strict interpretation of health-related codes (those having to do with discussions of one's physical and/or mental health) would be overly restrictive in terms of what respondents had to say about health-related matters, when 'health' is interpreted as having to do with well-being. In order to address this shortcoming, each respondent's file was scanned across all open-ended questions for other health-related commentary. The following questions proved to be particularly useful: "What helps you get through the rough times? What are some of the things you do to cope?"; "What is the best place you ever lived? What did you enjoy about it?"; "Do you have any advice for people who are homeless and looking for regular housing?"; and "What difference has it made for you to be in regular housing?" After this scan, we identified four additional sets of questions/topics where health-related themes were mentioned or implied. Health-related commentary was then organized for each of the respondent subgroups according to the six resulting themes (in addition to an 'other' category): (1) physical health, (2) mental health, (3) sense of control, (4) advice to others, (5) feelings and (6) particularly important services.

Table 5-4. Health-related quantitative results for immigrant and refugee respondents

Table 5-4. Health-related quantitative results for immigrant and refugee respondents

aF (1, 42) = 5.29, p < 0.05

23Among adults in families, those born in Canada were more likely than the others to report both physical and mental health problems:

She has fibromyalgia, asthma and bronchitis (cdn 214).
Post-traumatic stress disorder (from childhood experiences) and depression
(cdn 225).
Being bullied throughout my public school years affected my ability to concentrate on school work and get proper education. As a result, it disturbed me mentally until today
(cdn 530).

24No immigrants mentioned physical or mental health difficulties, and the reasons given by refugees for their health-related problems were specifically linked to their experiences of war and conflict:

The torture was worse than the war. War is just hearing the Boom Boom. It is not like having someone hold a gun to your head and your neck (ref 124).
I have shrapnel in my leg that prevents me from climbing stairs if an emergency occurs
(ref 524).

25The experiences of access to health care services were somewhat more similar. All of the adults in families reported difficulties:

I didn't get a doctor at Cobourg because they were not taking new patients (cdn 519).
Why did you not get care? Transportation; couldn't get to the office; not convenient; single dad (cdn 530).
Why did you not get care? Six months ago I tried to see a gp and made the appointment but the doctor had to cancel (ref 545).
Why did you not get care? When living with my husband, needed somebody to talk to and listen to—and nobody was there for me (imm 551).

26Overall though, the adults in families did not emphasize their health status or access to health services as primary factors that explained their homelessness, although health was sometimes used to explain why they didn't have sufficient income:

I had a mechanical back disorder, and if I wouldn't have that I would be working and making more money to avoid homelessness (cdn 521).
Due to my physical disability I can't obtain employment and that forced me to be on social assistance. As a result, the money doesn't allow me to rent a private accommodation
(ref 524).

27Access to affordable housing (and often, escape from abuse) were the factors that generated the most response. In fact, many of the adults identified moving into the family shelter as a strategic decision:

I was a single mother who is working and not making enough money. For a year, 85% of my earnings was going to my landlord. I couldn't continue doing that, and decided to come into the family shelter so I can find an affordable place to live. Before [that] I was living in a tent on Bank Street... (cdn 518).
I had been in an abusive relationship for 15 years and finally decided to leave, but I couldn't find an affordable place to rent and sought help in this shelter
(cdn 523).
I was in an abusive relationship for a long time. I couldn't accept that type of abuse any longer, and sought refuge in this shelter
(imm 527).
I have been in this shelter for 7 months, waiting to find an affordable place to live
(imm 509).
My two sons and I were renting a one-bedroom apartment and paid $818 for rent, while our income was $1162 from Ontario Works. In order to avoid being homeless, I had to take a portion of our food allowance... to pay the rent
(ref 524).

28For all of the adults in family respondents, the services that were highlighted as being of most benefit were closely tied to access to economic and/or housing resources. When health services were mentioned, it was typically as part of a broad range of services that together were perceived as helping the individual to become self-sufficient:

How do they help you? Welfare (Ontario Works) shelter, Native Women's group—came with me to Ontario Works, helped me get to Ottawa and into the shelter (cdn 139).
The shelter when I had no place to stay and the food banks when I had no food
(cdn 537).
How do they help you? When we have appointments: babysit. Worker gives what I need. If I leave a message... he helps right away, he does job well (imm 229).
The shelter for housing; legal aid for immigration; the community health centre for health services
(ref 147).

29In contrast to the adults in families, there were clearer distinctions among the single adult women, with Canadian-born women being much more likely than those born elsewhere to focus on health-related problems and services:

Self-mutilated. Bipolar disorder. Panic attacks (can 109).
Alcohol addiction
(cdn 240).
I can't work—my lower back has bad pain, neck muscle spasms go all the way down my arms
(cdn 112).
I need some nursing care because of my legs. I don't know if I will need a second amputation...
(cdn 149).
Seeing my family doctor at Centretown Community Health Centre
(cdn 149).
Do you receive help from outreach workers?
act Team— bring medication, my dad gives them $150/month, and they bring me $5/day; find out about housing; book doctor app'ts, drive me around, take me for coffee (cdn 101).

30The foreign-born adult women also mentioned physical and mental health problems, but their overall focus was more similar to the adults in families than they were to their single Canadian peers. For example, when asked what advice they would give to others in similar circumstances, the foreign-born women emphasized strategies that would help them exit homelessness, whereas the focus of the Canadian-born women was more on personal healing:

When they are in shelters they should get help and education—counselling, help for employment, getting a house—fill up everyday with activities. They have to follow you, push you to go out and look for work, housing... (imm 144).
You have to fight for a living and it will make you feel good; you need to work at it and never give up; there's always help
(ref 141).
Find counselling. Work things out with a counsellor. It's mostly addictions to drugs or alcohol or mental illness why people are homeless, they need to get help
(cdn 137).
Keep their appointments. Keep your act together. Don't fuck up
(cdn 240).

31The foreign-born single women and adults in families also were similar in their responses to questions about the difference it has made to be in regular housing (that they perceived as being of good quality):

What difference has it made for you to be in regular housing?
Big time difference. I am much better here (single adult, ref 901).
I am really happy to be provided with subsidized housing. And for the social assistance that the Government provides me. English courses are free of charge. Legal aid was a turning point to stabilizing my family situation. I know for sure in Ukraine they have no social network like this (single adult, imm 222).
What difference has it made for you to be in regular housing?
Peace of mind. My children have become healthy kids again after residing at this address (adult in family, cdn 517).

Discussion

32This study contributes to knowledge about the health of immigrants and refugees in Canada who have been homeless. In the analysis discussed here, the mental and physical health status of foreign-born respondents was higher than that of their matched, Canadian-born peers. But contrary to other research on immigrants' health, there was no evidence of a shrinking gap between the two groups over the two-year study period. Mental health status generally remained lower than that of the US normative sample but it did improve for all of the respondents over time. Immigrants and refugees' physical health status also was better than that of the Canadian-born respondents and, somewhat surprisingly, remained at a higher level than the US normative general population sample.

33It is plausible that improvements in mental health were associated with respondents who were in stable housing at follow-up and who perceived their housing to be of good quality. The adult-in-family respondents (both Canadian and foreign-born) were almost all stably housed at follow-up. Statistical analyses of the full Panel Study sample revealed that improvements in mental health status occurred when respondents perceived their living circumstances as being of good quality (Aubry et al. 2007). In the specific case of refugees, it is also plausible that feelings of safety and security had improved as compared to their situations at the beginning of the study.

34With regard to physical health status, the substantially more extensive reporting of chronic physical health conditions among the Canadian-born certainly provides one window on why there is a gap between the physical health status of those born in Canada and those born elsewhere. Although it is unsurprising that immigrants would have many fewer chronic conditions, given the extensive screening of their health status as a condition of entry, the case of refugees is more surprising, since their entry is not governed by the same health-related limitations. However, the superior health of refugees may be a testament to their resilience in surviving extremely difficult life circumstances in their home country and finding a way to escape these circumstances to a new country. Overall, our findings do provide further evidence of the healthy immigrant effect at least in terms of physical health among foreign-born adults who are homeless. Moreover, there was no evidence of a diminishment in physical health functioning among these individuals over the course of the two-year study.

35The health service utilization results do suggest a move towards similar rates of use among foreign-born and Canadian-born respondents over time, but this trend raises at least as many questions as answers. Given that foreign-born respondents report better mental and physical health, are they seeking the same kinds of health services as their Canadian-born peers or are they seeking help for settlement-related issues? The higher utilization of religious leaders and similarities in the rates that they and the Canadian-born respondents report in not getting the help they needed suggest that settlement-related issues are of particular concern for those born outside the country.

36The respondents' qualitative commentary revealed more similarities than differences among foreign-born and Canadian-born adults with family, but they also revealed clear distinctions between the foreign-born single women and their Canadian peers. It is likely that site-specific factors had a role to play. Among the full Panel Study sample, fully 97 percent of the adult-in-family group was stably housed at follow-up, and, of that group, 78 percent were living in subsidized housing. Given that homeless adults with children were placed on the priority list for subsidized housing in Ottawa, this trend is, at least in part, the unsurprising result of a municipal public policy choice. Among adult women, the probability of access to subsidized housing was considerably lower and depended on additional conditions such as mental health impairment or escape from domestic abuse: 73 percent were stably housed at follow-up, and, of that group, 51 percent were living in subsidized housing. Although respondents would not have been privy to the statistics presented here, the adult-in-family group might well have been aware of others who were successful in accessing subsidized housing, making the choice to move to the shelter a meaningful strategy in the search for affordable housing. The logic of using such a strategy in the case of single women would have been much less straightforward. In the case of foreign-born women, similar to the case of adults with family, it is likely that their poverty and/or domestic abuse were the driving factors in the decision to move to a shelter, whereas for the Canadianborn women, the reasons were more likely to do with interactions between mental and physical health limitations, domestic conflict and abuse and poverty.

Strengths and Limitations

37This study has several strengths. One is the Panel Study's site-specific success in re-interviewing 255 participants two years after first meeting them at an emergency shelter or drop-in in Ottawa. The Panel Study's interest in diversity and its systematic approach to capturing that diversity in a statistically rigorous manner is a second positive feature. It permitted the investigation of similarities and differences in the health status and health trajectories of matched samples of Canadian-born and foreign-born respondents living in Ottawa. The benefits of combining close-ended and open-ended questions in the interview contributed further insights into the 'healthy immigrant effect' for persons who have been homeless, and it identified additional reasons for examining the impacts of settlement-related concerns. The use of cultural interpreters and the ability to incorporate the experiences of both immigrants and those who arrived in Canada as refugees are additional significant features of the study.

38Our study also had a number of limitations that need to be taken into account in interpreting the findings. The first was that the study was not representative of the overall homeless population in Ottawa; in addition, the sample did not include single adults and adults in families who were living on the street or temporarily living with friends or family. A further limitation of our sampling strategy was the reliance on shelter staff to find participants in the single adult and adult-in-families subgroups who matched specific characteristics. This process had the potential to introduce some bias toward sampling higher functioning individuals. There were also refusals by some of the individuals invited to participate. Another limitation was the level of attrition in the study, as 38 percent of the original participants were lost at follow-up. There were no differences found between participants and non-participants in the follow-up interview on all of the compared characteristics with the exception of length of residency in Ottawa (Aubry et al. 2007).

39All of the information collected in the study was self-report in nature. Self-report information may be prone to inaccuracy because of faulty memory, lack of information or discomfort with self-disclosure. Related to the use of self-report measures in the Panel Study, it is important to note that the sf-36 provides a subjective assessment of physical health functioning and mental health functioning. Other limitations associated with the use of sf-36 include the potential lack of relevancy of some of the items in the measure related to inquiring about physical functioning and social functioning from people who are experiencing homelessness. As well, our use of cultural interpreters may have affected the interpretation and response of participants on some of the items. Despite these limitations, scores on the sf-36 showed the expected differences between immigrants and refugees and Canadian-born participants, with immigrants and refugees reporting better physical health functioning and mental health functioning. As well, the measure proved to be sensitive to capturing improvements in mental health functioning over time for both groups of participants.

40Another limitation of our study was the relatively short length of the follow-up period (i.e., two years). It was of sufficient length to capture the housing instability experienced by a relatively large proportion of our participants. However, a longer period of time may have shown more changes in the health-related characteristics. A further limitation was the fact that no distinctions were made among the foreign-born respondents who were recent arrivals and those who had been in Canada for a longer period of time. This shortcoming may have compromised the ability to examine the healthy immigrant effect over time. A final limitation was the relatively small numbers of respondents that were compared based on matched samples.

Conclusions

41Beiser's (2005) review of extant research on the health of immigrants and refugees revealed a complex picture and lent support to his conclusion that "unexpected and paradoxical findings underline the need to take account of heterogeneity in future studies . . ." (539). The results reported here lend support to his conclusions and also raise questions about the impact of site-specific public policy measures on respondents' health status and health services utilization. Although the study results do provide support for the widely observed healthy immigrant effect, the similarities noted between immigrants and refugees raise questions about its causal roots. So too do the observations about similarities between the foreign-born and Canadian-born adults with families. These findings have implications for policy-makers, health administrators and health and social care professionals. This study suggests that the existence of a certain difference (such as country of origin) does not automatically mean a linear relationship with other distinctions (such as reasons for being homeless). In fact, our data suggest that adults with children, whether born in Canada or elsewhere, are more similar than different, whereas this is not the case among unaccompanied adult women. The data also raise further questions about the circumstances under which a healthy immigrant effect occurs and suggests that contextual factors need careful consideration before inferences are drawn on the basis of this tendency. Thus, in the case of Chui and colleagues (2009), the focus on English-speaking immigrants in Toronto homeless shelters needs to be taken into account in drawing inferences from that study. It would also be useful to investigate healthy immigrant effects among persons who have been homeless in locations other than Toronto or Ottawa.

42It is clear that further research is required and that its ideal form would be both longitudinal and multi-site. The Housing and Health in Transition study, now underway in Toronto, Vancouver and Ottawa, has the potential to contribute further insights about the health of single adults, both foreignand Canadian-born, who have been homeless or unstably housed in these three cities.

Bibliographie

References

Ali, J. S., S. McDermott and R. G. Gravel. 2004. "Recent Research on Immigrant Health from Statistics Canada's Population Surveys." Canadian Journal of Public Health, 95(3): I9–13.

Argeseanu Cunningham, S., J. D. Ruben and K. M. Narayan. 2008. "Health of Foreign-Born People in the United States: A Review." Health & Place, 14(4): 623–35.

Aubry, T., F. Klodawsky, E. Hay and S. Birnie. 2003. Panel Study on Persons Who Are Homeless in Ottawa: Phase 1 Results. Ottawa: Centre for Research on Community Services.

Aubry, T., F. Klodawsky, R. Nemiroff, S. Birnie and C. Bonetta. 2007. Panel Study on Persons Who Are Homeless in Ottawa: Phase 2 Results. Ottawa: Centre for Research on Community Services.

Aubry, T., F. Klodawsky, E. Hay, R. Nemiroff and S. Hyman. 2004. Developing a Methodology for Tracking Persons Who are Homeless over Time. Final Report. Ottawa: Canada Mortgage and Housing Corporation.

Beiser, M. 2005. "The Health of Immigrants and Refugees in Canada." Canadian Journal of Public Health, 96 (March/April): S30–S44.

Castro, F. G. 2008. "Personal and Ecological Contexts for Understanding the Health of Immigrants." American Journal of Public Health, 98(11): 1933.

Chan, A., E. Pristach and J. Welte. 1994. "Detection by the cage of Alcoholism or Heavy Drinking in Primary Care Outpatients and the General Population." Journal of Substance Abuse, 6: 123–35.

Chui, S., D. Redelmeier, G. Tolomiczenko, A. Kiss and S. Hwang. 2009. "The Health of Homeless Immigrants." Journal of Epidemiology and Community Health, 63: 943–48.

cfo (Community Foundation of Ottawa). 2009. Ottawa's Vital Signs 2009: The City's Annual Check-Up. Ottawa: Community Foundation of Ottawa.

Dey, A. N. and J. W. Lucas. 2006. "Physical and Mental Health Characteristics of U.S. and Foreign-Born Adults: United States, 1998–2003." Advance Data, (369): 1–19.

Dunn, J. and I. Dyck. 2000. "Social Determinants of Health in Canada's Immigrant Population: Results from the National Population Health Survey." Social Science & Medicine, 51: 1573–93.

Farrell, S., T. Aubry and E. Reissing. 2002. "Street Needs Assessment: An Investigation of the Characteristics and Service Needs of Persons Who Are Homeless and Not Currently Using Emergency Shelters in Ottawa." [on-line]. Homeless Hub. http://www.homelesshub.ca/Library/Street-Needs-Assessment-An-Investigation-of-the-Characteristics-and-Service-Needs-of-Persons-who-are-Homeless-and-not-Currently-using-Emergency-Shelters-in-Ottawa-54361.aspx [consulted January 12, 2014].

Fennelly, K. 2007. "The 'Healthy Migrant' Effect." Minnesota Medicine, 90(3): 51–53.

Frankish, C., S. Hwang and D. Quantz. 2005. "Homelessness and Health in Canada: Research Lessons and Priorities." Canadian Journal of Public Health, 96(March/April): S23–S29.

Kappel Ramji Consulting Group. 2002. Common Occurrence: The Impact of Homelessness on Women's Health: Phase II: Community Based Action Research: Final Report Executive Summary. Toronto: Sistering: A Woman's Place.

Klodawsky, F., T. Aubry, B. Behnia, C. Nicholson and M. Young. 2007. "Comparing Foreign-Born and Canadian-Born Respondents to the Panel Study on Homelessness (Phase 1)." Our Diverse Cities: Ontario, 4(Fall). Ottawa: Metropolis, 51–53.

——. 2005. The Panel Study on Homelessness: Secondary Data Analysis of Respondents Whose Country of Origin Is Not Canada. Ottawa: Ottawa: National Homelessness Secretariat.

Huh, J., J. A. Prause and C. D. Dooley. 2008. "The Impact of Nativity on Chronic Diseases, Self-Rated Health and Comorbidity Status of Asian and Hispanic Immigrants." Journal of Immigrant and Minority Health, 10(2): 103–18.

Mayfield, D., G. McLeod, P. Hall. 1974. "The cage Questionnaire: Validation of a New Alcoholism Screening Instrument." American Journal of Psychiatry, 131(10): 1121–23.

McDonald, J. T. and S. Kennedy. 2004. "Insights into the 'Healthy Immigrant Effect': Health Status and Health Service Use of Immigrants to Canada." Social Science & Medicine, 59(8): 1613–27.

Newbold, B. 2005. "Health Status and Health Care of Immigrants in Canada: A Longitudinal Analysis." Journal of Health Services & Research Policy, 10(2): 77–83.

Paradis, E., S. Novac, M. Sarty, J. Hulchanski. 2008. Better Off in a Shelter? A Year of Homelessness & Housing among Status Immigrant, Non-Status Migrant, & Canadian-Born Families (Research Paper 213). Toronto: Centre for Urban and Community Studies, University of Toronto.

Region of Ottawa Carleton. 1999. Homelessnesss in Ottawa-Carleton. Ottawa: Region of Ottawa Carleton.

Simich, L., M. Beiser, M. Stewart and E. Mwakarimba. 2005. "Providing Social Support for Immigrants and Refugees in Canada: Challenges and Directions." Journal of Immigrant Health, 7(4): 259–68.

Singh, G. K. and R. A. Hiatt. 2006. "Trends and Disparities in Socioeconomic and Behavioural Characteristics, Life Expectancy, and Cause-Specific Mortality of Native-Born and Foreign-Born Populations in the United States, 1979–2003." International Journal of Epidemiology, 35(4): 903–19.

Skinner, H. 1982. "The Drug Abuse Screening Test." Addictive Behaviour, 7(4): 363–71.

Susser, E., R. Moore and B. Link. 1993. "Risk Factors for Homelessness." American Journal of Epidemiology, 15: 546–56.

Ware, J., M. Kosinski and B. Gandek. 2002. sf-36 Health Survey: Manual and Interpretation Guide. Lincoln, RI: Quality Metric Incorporated.

Notes

1 Standardized scores involve converting the raw score to a scaled score based on a normative sample of the 1998 general US population (ages 18–64; n = 6742) where the mean is 50 and the standard deviation is 10.

Table des illustrations

Titre Table 5-1. Overview of panel study subgroups, sampling criteria, number of recruitment sites and baseline and follow-up sample size
Légende *Youth were recruited at shelters and drop-in facilities in equal numbers
URL http://books.openedition.org/uop/docannexe/image/785/img-1.jpg
Fichier image/jpeg, 116k
Titre Table 5-3. Health-related quantitative results for panel study matched Canadian-born and foreign-born respondents, and immigrant and refugee respondents
Légende aF(Group) (1, 88) = 6.79, p < 0.02; bF(Time) (1, 88) = 8.06, p < 0.01; cF(Group) (1, 88) = 9.01, p < 0.005;dF(Group) (1, 90) = 22.01, p < 0.001; eF(Group) (1, 87) = 7.41, p < 0.01; fChi-square (df = 1) = 4.09, p < 0.05
URL http://books.openedition.org/uop/docannexe/image/785/img-2.jpg
Fichier image/jpeg, 156k
Titre Table 5-4. Health-related quantitative results for immigrant and refugee respondents
Légende aF (1, 42) = 5.29, p < 0.05
URL http://books.openedition.org/uop/docannexe/image/785/img-3.jpg
Fichier image/jpeg, 155k

Auteurs

Professor in the Department of Geography and Environmental Studies at Carleton University. Her areas of expertise include public policy and social inclusion/exclusion in cities and feminist perspectives on cities, community organizing, housing and homelessness. She was the Co-Principal Investigator of the Panel Study on Homelessness in Ottawa, the first large-scale longitudinal survey of homeless persons in Canada. Her work uses both quantitative and qualitative methods within a collaborative, community-based framework

Professor in the School of Psychology and Director of the Centre for Research on Educational and Community Services at the University of Ottawa. Dr. Aubry’s research is focused in the areas of community mental health, homelessness and program evaluation of health and social services for marginalized populations. He is member of the National Research Team and the Co-Lead of the At Home/Chez Moi research demonstration project in Moncton and a Co-Principal Investigator of the Homelessness and Health in Transition study

Earned her Clinical psychology from the University of Ottawa in 2010. Her doctoral research focused on the community integration of women who have experienced homelessness. She currently works as a school and clinical psychologist in Gatineau, Quebec and is an Associate of the Centre for Treatment of Sexual Abuse and Childhood Trauma in Ottawa

© Les Presses de l’Université d’Ottawa | University of Ottawa Press, 2014

Conditions d’utilisation : http://www.openedition.org/6540