Abstract
Underlying the daily lives of people with experiences of homelessness and mental illness is a complex interplay of individual and structural factors that perpetuate cycles of inequity. The introduction of novel methodological combinations within qualitative research has the potential to advance knowledge regarding the experience of health equity by such individuals and to clarify the relationship between these experiences and broader structural inequities. To explore the lived experience of inequity, we present a thematic analysis of narrative interviews in conjunction with timelines from 31 adults experiencing homelessness and mental illness. Use of these methods together enabled a novel and expanded appreciation for the varied ways in which differential access to the social determinants of health influences the trajectories and experiences of inequity for people who are homeless and mentally ill. The further utility of these methods for better understanding the experience of inequity is explored and implications for research, policy, and practice are discussed.
Introduction
Equal access to the resources and benefits of society and “a life one has reason to value” (Marmot, 2007) are widely held as key values of liberal democracy. However, a strong social gradient in health runs across most societies, and where we are placed in relation to other people affects access to various socio-economic opportunities (Wilkinson & Pickett, 2006). Individuals experiencing homelessness and mental illness occupy a position near the bottom of this social gradient. Underlying the experiences of such individuals is a complex interplay of individual and structural factors that perpetuate cycles of inequity (Cronley, 2010). While most research on homelessness has focused on the exigencies of daily living, including mental health and substance use issues, physical illness, and mortality (see Hwang, Wilkins, Tjepkema, O'Campo, & Dunn, 2009), many researchers have stressed the importance of understanding the role of social and structural inequities, particularly as they relate to individual health and public systems of care (Shinn, 2007; Snow, Anderson & Koegel, 1994). Nevertheless, few methodologies exist to bridge and integrate these micro- and macro-level perspectives (VanderStaay, 1994).
This paper begins with a brief description of the interrelated concepts of health equity, social justice, and the social determinants of health (SDoH), with a particular focus on adults experiencing homelessness. We review evidence related to the use of qualitative methods to promote knowledge of individual experiences of inequity in relation to broader structural factors. We then present a rationale for using narrative interviewing and visual timeline methods together, and suggest how they can be used to integrate micro- and macro-level perspectives to advance research on health equity for individuals experiencing homelessness and mental illness.
Inter-Related Concepts: Health Equity, Social Justice, and the Social Determinants of Health
Health equity is presented in the literature in different ways, including the fair distribution of quantifiable benefits and burdens, equal access to the SDoH, and through an appreciation of the nature and cycle of oppression (Varcoe, Pauly, & Laliberte, 2011). The SDoH are central to the concept of health equity because the distribution of important social opportunities amongst individuals and groups in society is shaped at a broader structural level by historical, social, economic, and political factors (Raphael, 2008). Within the Canadian context, ten key determinants have been identified including employment security, working conditions, equitable income distribution, food security, early childhood development, education, healthcare services, adequate housing, social safety nets, and social inclusion (Mikkonen & Raphael, 2010). Thus, health equity can be defined as “the absence of systematic and potentially remediable differences in one or more aspects of health across socially, demographically, or geographically defined populations or population subgroups” (Starfield, 2006, p.13).
The lack of access to the SDoH is obvious among individuals experiencing homelessness. In Canada, homelessness is growing, with annual street counts in some cities seeing triple-digit increases over the past decade (Laird, 2007). Over recent decades, structural changes to the economy and shifts in government policy have led to reductions in income support, affordable housing stock, and funding for residential and community-based care for individuals with severe mental illness (Brzozowski, Gervais, Klein, & Suzuki, 2010; Gaetz, 2010). The impacts of these and other policies have played decisive roles in perpetuating vulnerability to homelessness for an increasingly diverse population of Canadians.
The concept of health equity is closely related to that of social justice. Both concepts rest on the assumption that all members of society are valued equally and are worthy of access to and receipt of the benefits of society (Klowdawsky, 2009). Some researchers have argued for a shift from distributive social justice (focused on income) to a more relational concept based on fair access to social goods such as rights, opportunities, power, and self-respect (Kenny, Sherwin, & Baylis, 2010). Others call for the inclusion of multiple dimensions, including political, economic, and cultural factors (e.g., Fraser, 2007). Both relational and broader frameworks are important because complex social processes, including marginalization and exclusion, are constructed by and embedded in personal trajectories of stress and coping as well as historical, political, and economic contexts. Nevertheless, the interaction of these structural factors with individual life course trajectories is not well understood. There is a need to expand the empirical methods used to examine the intersections of personal experience and the social determinants known to affect health equity.
Health Equity and the Need to Advance Research Methods
Homelessness research has largely been dominated by the use of survey data and other methods that focus on the individual rather than the underlying structural factors that influence how groups differentially participate in public life (Snow, Anderson, & Koegel, 1994). This primary focus on the individual has increased our understanding of risk factors for homelessness and various adverse outcomes; however, it fails to capture the complex social structures that contribute to income disparity, poverty and ultimately homelessness. Even qualitative methods, which have traditionally been concerned with viewing participants in a broader context, often focus on individual experiences of deprivation (VanderStaay, 1994).
A few researchers concerned with health equity among homeless populations have attempted to bridge micro- and macro-level perspectives by utilizing qualitative methods. For example, Shibusawa and Padgett (2009) used narrative interviews in combination with case studies to focus on the “person-in-context” for older individuals with experiences of homelessness and chronic mental illness. This combination of methods draws attention to the ways in which the participants' lives were profoundly affected by a lack of community-based services, a shortage of affordable housing, and the unintended policy consequences of treating mental illness and addiction in silos. Hodgetts, Radley, Chamberlain, and Hodgetts (2007) examined relational aspects of homelessness and its health consequences through the use of photo-elicitation interviews with 12 rough sleepers in London. This study highlighted the relevance of “embodied deprivation” and how social inequalities manifest as health inequities. Finally, Finley and Diversi (2010) recently used a narrative analysis to examine homelessness as a personal experience and as a public issue. The authors highlight dominant public narratives of homelessness and poverty as individual trouble and contrast these with the concept of homelessness as a “systemic expression of ideologies of domination (p. 7).” While these and other studies (e.g., Drury, 2003; Gerson, 2007; Marshall, Kerr, Shoveller, Montaner, & Wood, 2009; Wasserman & Clair, 2011) have examined the link between the individual and structural factors related to homelessness, very little research has examined the benefits of combining qualitative methods.
Combining Research Methods
Specific research methods–the practical activities of research–stem from epistemological and methodological choices (Carter & Little, 2007). In this paper, we assume that knowledge is jointly constructed in collaboration with participants in a specific context. Furthermore, while we attempt to identify key themes and issues that contribute to trajectories of social and health inequities and homelessness, we do not believe these are static, measurable entities. We hope that these issues will become apparent through the stories participants tell in collaboration with the interviewers. As described by Connelly and Clandinin (2006), we believe that people interpret their experiences by creating stories that give personal meaning to their daily lives. Thinking narratively, we attend to various aspects of storytelling including temporal dimensions (past, present, future), social dimensions (inner and outer, personal and structural), and place (physical and social spaces) (see Clandinin, Murphy, Huber, & Orr, 2010).
Research methodologies and methods are rarely “pure,” and disciplinary understanding along with experience provides a basis for meaningfully combining or modifying existing methodologies. In an effort to bridge micro- and macro-level perspectives on homeless individuals' experiences of inequities, we conducted narrative interviews and developed corresponding personal timelines for each participant.
Narrative Interviewing
Narrative interviewing allows participants to engage in an evolving conversation wherein interviewer and interviewee collaboratively produce and make meaning of the events and experiences reported (Mishler, 1995). The participant is invited to tell their story, for example, how they became homeless. However, experience always overrides its description; events may be briefly noted and given little significance. Often, further questioning is required to help participants recall the details, turning points and other shifts in their thoughts, emotions, and actions (Reissman, 2008). In general, less structure in the interview guide gives greater control to both interviewer and interviewee to jointly construct narratives. Among researchers, there is considerable variation in definitions of personal narrative, often linked to discipline. We follow the tradition common in psychology and sociology wherein personal narratives entail extended accounts of lived experience in a particular context (see Mishler, 1995). Innovations in narrative interviewing include combining observation, embedded relationships, multiple conversations over time, as well as visual data. However, to our knowledge, no one has explored combining narrative interviewing with timelines among homeless populations.
Personal Timelines
The construction of timelines is one method of visually organizing rich narrative data. Timelines are generally constructed by highlighting events of interest in an individual's life in chronological order. The significance and meaning attached to events may also be shown. Timelines facilitate recollection and sequencing of personal events and they are useful for comparison with other data to confirm or complete a life history or to place a particular research construct or clinical issue in the context of other events (Gramling & Carr, 2004). Sometimes timelines for a group of individuals are aggregated to aid the detection of patterns and sequences over time (e.g., Berends, 2011). For example, one could examine the pattern of housing, vocational and social experiences that precede absolute homelessness. The timeline method is underutilized within qualitative research on health and homelessness, as the majority of qualitative data collection utilizes the interview alone (Harris & Huntington, 2001; Rhodes & Fitzgerald, 2006).
There are obvious challenges in representing life stories in an abbreviated manner such as the timeline. Summarizing and quantifying narrative data risks reducing and oversimplifying people's stories (Boyd, Hill, Holmes, & Purnell, 1998). Thus, we propose that timelines are most effective when used in conjunction with a more complete data source such as narrative interviews. In combination, these methods can enhance understanding of trajectories of resilience and risk and assist in identifying points for intervention.
Methods and Research Context
Sampling and Recruitment
The data for this study were collected as part of the Vancouver At Home study, a nationally funded intervention project for adults experiencing homelessness and mental illness. Eligibility criteria include legal adult status (over 19 years of age), presence of a current mental disorder, and being absolutely homeless or precariously housed (see Goering et al., 2011). Participants were recruited through referral from a wide variety of agencies that serve the homeless including shelters, drop-in centres, outreach teams, inpatient hospital wards, and criminal justice programs. All participants met with a trained research interviewer who explained procedures, confirmed study eligibility, and obtained informed consent.
Participants in the present sub-study were both randomly and purposively selected from the larger sample of Vancouver At Home study participants. Purposive sampling was based on both typical and unusual cases, in an effort to represent differences across gender, ethnicity, sexual orientation, duration of homelessness, and degree of functional impairment. Within one month of enrollment in the larger study, selected participants were contacted and asked if they would like to participate in a “personal story interview.” Participation was voluntary, and two out of 33 participants declined to participate. The current study is based on interviews from a subsample of 31 participants who were selected from the first 400 participants enrolled in the Vancouver site from the beginning of recruitment in October 2009 to December 2010.
Data Collection
Three university-based researchers and one peer interviewer, who had experienced homelessness, conducted interviews with study participants. Interviews lasted from one to two hours and were conducted at a setting chosen by the participant, usually a shelter or the institution where they were staying prior to contact with the project. All participants gave informed consent and received $30 CAD upon completion of the interview. The research ethics board at the authors' university approved the research protocol.
Using a semi-structured interview format, participants worked with interviewers to co-construct a personal story highlighting (a) their pathway into homelessness; (b) experiences of being homeless or inadequately housed; (c) experiences around first learning they had a mental illness and obtaining help for their illness; and (d) high, low, and turning points in their life. Interviews were audio recorded and transcribed verbatim.
Data Analysis
According to Frank (2000), people tell stories but narratives come from the analysis of stories. Therefore, the researcher's role is to interpret the stories in order to uncover the underlying narrative that the interviewee may not be able to voice themselves. Coding not only functions to make concepts logistically manageable, it also brings concepts into focus when they otherwise would remain diffuse in the narrative structure of the data. In this study, thematic analysis was used to examine the interview transcripts. To ensure rigor, the research team met four times during the early phase of thematic analysis to co-code and discuss emergent themes in the narratives. Subsequently, two researchers coded all transcripts line-by-line and independently identified repeated or similar codes to build a set of overarching themes (Morse & Field, 1995; Strauss, 1987). After a thorough review of the transcripts and other supportive documents (e.g., field notes), conceptual impressions were integrated into key thematic areas. At this point, thematic areas and initial interpretations were shared with field interviewers and researchers whose interpretations were cross-checked with the initial findings.
According to this approach, emphasis is on the content of the text (what is said) rather than on structural or discourse analysis (how it is said). The thematic approach is useful for finding common thematic elements across participants and the events and experiences they report. Our thematic analysis reflects both ideas brought to the data set from the research questions and the SDoH as an organizing framework (i.e., top-down) as well as being open to new themes that emerge in the data (i.e., bottom-up). For example, we looked for instances in which participants described life events that were clearly related to access to the SDoH (e.g., housing, healthcare, employment), and for scenarios during which a participant stated explicitly that they were being treated unfairly. This framework helped us consider health and social inequities at various levels of abstraction (e.g., individual, community, larger systems), yielding a method for linking concepts within and between levels of scale encountered in the coding and categorization process.
Following thematic analysis of the interviews, timelines were constructed for each individual interview to visualize trajectories of events and experiences in the context of relevant personal, historical, and political events. The assembly of each timeline began with outlining major developmental stages and adding key life transitions (e.g., leaving the parents' home, marriage, birth of a child) as well as key life events and stressors (e.g., initiation of drug use, imprisonment, housing and vocational changes, loss of loved ones). When available, we made note of the meaning and significance key events held for the individual. Next we reviewed each individual timeline and identified key SDoH to illustrate common patterns related to access and at what developmental stage particular determinants were most relevant. Analysis of the timelines focused on the timing and sequence of various events within the context of broader structural factors. It should be noted that these timelines were not constructed in partnership with participants but were used as an analytic strategy after the narrative interviews had been completed.
Finally, an aggregated timeline was created to highlight overarching patterns of inequity for this group. Two researchers independently reviewed the individual timelines and identified the key events, sequences of events, and themes for each developmental stage (e.g., childhood, adolescence, early adulthood, later adulthood). Discrepancies were discussed within the larger research team until consensus was reached. Frequently occurring events and themes were then grouped by developmental stage to form an overall timeline. We hope the use of different qualitative methods can forge connections between personal stories and social structure, that is, both the personal and the political.
Research Context
Socio-historical events that have shaped Vancouver's Downtown Eastside (DTES) set the stage for the personal narratives discussed in this study, as most participants moved to the DTES after becoming increasingly marginalized from mainstream society. As one of few neighbourhoods in the city that had affordable housing, mostly in the form of rooming houses, many low-income people have been drawn to the DTES over the past 50 years. The 1980s marked the onset of rapid gentrification throughout the city as well as the arrival of an inner-city illicit drug market, and the transition of psychiatric care from institutions to the community.
Like all Canadian cities, Vancouver has been greatly affected by the federal reduction in affordable housing. In addition, there has been a sharp decline in private-market single room occupancy (SRO) accommodation across the province. Existing buildings are declining at a steep rate due to redevelopment and the demolition of increasingly dilapidated stock. Moreover, in many communities, adequate and affordable housing is beyond the means of people who rely on income support. Therefore, health and social inequities among people experiencing homelessness in Vancouver's DTES are shaped by a complex interplay of social, political, economic, and historical factors.
Results
Sample Characteristics
Demographic characteristics of the participants are presented in Table 1. Age ranged from 26 to 66 years (mean = 45 years). The sample included 19 males (61%) and 11 females (35%). Over 50% of the participants were White and 26% were Aboriginal. Most participants were single, never married (55%) or divorced, separated, or widowed (32%). The vast majority (81%) of participants were absolutely homeless, while the remainder were precariously housed. At the time of the interview, 42% had attended but not completed high school. The categories of mental disorders that were most frequently identified in the sample were Psychotic Disorder (45%), Major Depressive Episode (42%), and Mood Disorder with Psychotic Features (32%). In addition, 68% met criteria for current Substance Dependence and 19% for Alcohol Dependence. Participants' main source of income was social assistance, with 48% of participants receiving basic income assistance, and another 48% receiving support associated with a disability.
Demographic Characteristics (n=31)
Key Themes from Narrative Interviews
Based on our analysis of the narrative interview data, the following key themes emerged: longstanding social devaluation, feeling trapped (arriving at “the end of the road”), and experiencing a profound lack of autonomy or the illusion of choice.
Theme 1: Longstanding Social Devaluation
Aware of their status as a “homeless person,” many participants reflected on how they were unfairly treated by public systems of care and by members of mainstream society. Susan 1 is a 30-year-old white woman who was living on the streets at the time of the interview. She was homeless for the first time at the age of 16 and gave birth to a baby girl in her early twenties. Reflecting on her experience of being pregnant, she states:
Like, when you're pregnant, it's treated as a disease. And all these, like, social workers come in when you're pregnant. God! Chaos! You know? And, it's like ‘Don't drink,’ you know? It's not like, ‘Wow! You're having a baby. Great! How can we help?’
As a homeless mother, Susan describes feeling judged for becoming pregnant and for using alcohol. Later in the interview, she states that she is “under a lot of scrutiny” and experiences constant fear that she will lose her child. Despite being judged by others, Susan has a strong desire to be treated with respect. Reflecting on what would make her feel a part of the community, she states:
Um, I think when I'm respected. When people don't judge me because of my mental illness or my poverty or my circumstances, you know? When people see me as an equal, like value my opinions. And when I'm able to give and they're able to give. It's a give and take… Where it's not like ‘we're going to take care of you, make you our project.’
Joe, a 54-year-old white man, reports being severely beaten as a child by his father and running away from home at the age of 9. In his early adulthood, he owned his own trucking business, was married for several years, and had a daughter. At the age of 44, a drunk driver killed his daughter. Joe began using drugs and his marriage ended, leading to a prolonged period of severe depression.
He also experienced his first episode of homelessness at the age of 44. He has been homeless for ten years, living primarily in the DTES. He states:
No landlord wants somebody living in their apartment using drugs. It's that simple. Unless, again, you're in skid row… I have to live in a predominantly social area. You know, like where people watch – they know their neighbours – like what I did when I was with my wife… This is what I want to do again. I want to have a lifestyle that I can be proud of, not something I've got to hide from.
Given his substance use, Joe has never felt accepted in neighbourhoods outside of the DTES. He feels he needs to hide his lifestyle from people and as a result feels ashamed and of lesser value. Social systems and structures often reinforce homeless individuals' sense of failure and inadequacy, and thus play an important role in creating and perpetuating social and health inequities.
Theme 2: Feeling Trapped
Many participants expressed feeling “trapped” in the DTES. For most, the neighbourhood represents the “end of the road,” a place one goes when there are no other options. Following the end of his marriage, Joe started using drugs more frequently and moved to the DTES. Despite his desire to relocate and change his lifestyle, he describes experiencing great difficulties trying to leave:
I've been down in the lower eastside for almost 10 years. In and out, in and out… The drugs follow you, no matter where you go. You're going to find them. The last three years was serious thinking about not living this lifestyle anymore… There's not enough housing in this area. You go looking for minimum $375 and all you're going to find is cockroaches and bedbugs and all that kind of stuff… I've tried it. I've lived in a couple of little apartments and – no – I need something clean. I need something out of the lower eastside. And it's hard to break out of there. Once you're down there it's hard to break out.
John also describes feeling trapped in a place that reinforces his lack of self-worth. He is a 49-year-old Black man who, at the time of the interview, had been homeless for 14 months. He has cycled between jail and living on the streets for over 15 years. He states:
And now it's like I go to a job and I can't do it. I gotta quit and that looks bad on me cause I'm Black. So I'm barely making anything. I've been trying to get into [social housing] since 2007… Everybody in the east end has come to hate me… It's like I'm stuck here. Wish I could just fall asleep and not wake up… I want out so bad. Here I'll always be an addict… There's no escape. It creeps into people.
Theme 3: Lack of Autonomy
Many participants were keenly aware that they lacked autonomy to make key decisions about their lives. Despite the hardship of life on the streets, some participants described a sense of freedom that they did not experience in institutions and shelters. Most participants were keenly aware that the choices they made were constrained and described “making the best” of a difficult situation. Wayne is a 51-year-old Aboriginal man who has experienced homelessness episodically for 22 years. In describing his choice to live on the street he states:
Yeah, I've stayed in those [SROs] but I'd rather stay on the street… I like my own space. My own bathroom, my own kitchen. I buy things for me not for everybody else…. I don't like too many bed bugs … I guess that's one thing about SROs is they're still trying to get, like, money from guests and I don't think that's right. I'd much rather stay on the streets.
While expressing that he would “rather stay on the streets,” Wayne is making this choice based on the fact that there are no adequate housing options available to him. For some, choice is influenced and constrained by past history, for example, being institutionalized in hospitals or prisons, which was commonly reported. For many, their history with the corrections system began at a very early age and continues. For example, Ian, a 26-year-old white man, describes being in jail since adolescence. Ian was adopted into a violent family as a young child and left high school in grade 9. He expresses thoughts surrounding the inevitability of returning to jail, no matter what choices he makes:
Mostly my future is probably being locked up, where I belong. I have warrants in [Province] and they want seven years federal. So, basically, it's go back there and go in for seven years.
[Interviewer: Yeah, do you think that's where you belong?]
I've been incarcerated since I was 12 years old. It's just a way of life. Three meals, clothes, and a bed and a shower. What more can you ask for…? Just going back to jail. It's just stability for me. It's not like being in a shelter or anything. At least going to jail I can get the help that I need, the medical attention I need. I can just basically get everything I need.
Lack of opportunity and constrained choices were also prominent for Alice, a 33-year-old white woman who left home at the age of 16 to escape severe abuse from her parents. She describes her first experience with subsidized housing and with trying to get the help she needed:
I spent my first GST cheque and bought a plane ticket on my own for $200 to Vancouver. I got here alone with nobody. And I had zero dollars, nowhere to live, no way of getting on welfare. It took me approximately seven months to get on welfare. First three months I was here, I was completely homeless. And then I ended up staying with somebody, got hooked up in the sex trade. For a place to stay, he gave me a rent receipt for three months so I could get onto assistance. So I had to sell my body just to get on assistance.
Many participants in our study are aware of the impact that adverse childhood experience, poverty, and lack of key social opportunities have had on their current position within society. In describing their pathways into homelessness and hopes for the future, many participants described trajectories of accumulating risk and marginalization that contributed to their current experience of social devaluation, despair, and constrained choices. It is noteworthy that many participants' reflections point to structural factors operating beyond their control. These findings underscore the complex interplay among micro- and macro-level contexts and highlight the fact that reduced opportunities for meaningful participation in society make it difficult for one to live “a life one has reason to value.”
Personal Timelines
As a means of organizing a large volume of rich and complex data, we created a timeline for each participant. Timelines focused on social and structural determinants of health and experiences of inequity. By developing these timelines, we were able to anchor key life events and experiences as they related to important developmental stages (e.g., early childhood, adolescence) and to examine cumulative adversity across the lifespan (Kubiak, 2005). Analysis of the timelines focused on the timing and sequence of various events within the context of broader structural factors.
Figure 1 illustrates a timeline for Clara, a 54-year-old white woman who had been absolutely homeless for two years and precariously housed for over 18 years. Her timeline demonstrates the presence of repeated abuse beginning in early childhood, early initiation of drug use and theft, and leaving school after Grade 8. This period represents a key opportunity for early intervention; however, no services were received by Clara at this time. In her early twenties, Clara became pregnant, was incarcerated for three years, and lost custody of her child. Upon release from prison, she regained custody of her child, got married, gave birth to another child, and abstained from substance use for 12 years. This marks a significant period of stability and resilience given her early trajectory. Separation from her husband and his subsequent death prompted a relapse into substance use and she lost custody of both of her children.
Timeline for Clara, a 54-Year-Old White Woman who Has Been Homeless
Typical Timeline for Individuals Experiencing Homelessness and Mental Illness (n=31)
Unable to draw on the resources of family and friends, Clara moved to Vancouver's DTES and became further marginalized. The timeline in conjunction with the narrative interview clearly illustrates a key turning point, at age 35, when her husband died and her drug use escalated resulting in the loss of her children. After losing her children for the second time, her sense of hope and willingness to persevere was severely compromised. Furthermore, her move to the DTES in 1992, a community where homelessness, poverty, and substance use were highly prevalent signaled “failure” in her eyes. The 1990s marked the peak in injection drug use and related harms in the DTES. It also marked a period where adequate and affordable housing became increasing harder to find in the neighbourhood.
When comparing individual timelines across our sample, despite diverse individual stories, we were struck by the common factors among participants' pathways into homelessness and experiences on the streets (see Figure 2). The majority of participants, despite age, gender, and ethnicity, described significant ruptures in primary attachment relationships during early childhood. Physical, sexual, and emotional abuse were commonly reported as well as frequent separations from caregivers due to domestic violence, parental hospitalization, and separation and divorce. The majority of participants grew up in families with limited financial and social resources and experienced housing and financial instability from a young age. The majority of participants lived with their parents or guardians until late adolescence and first experienced homelessness in their late twenties. Only three participants described stable family upbringings and became homeless later in life after a period of residential and financial stability.
Most participants traced symptoms of mental illness to middle childhood and difficult family and social environments they experienced at that time. Feelings of depression, anxiety, and anger were frequently reported, as well as problems with parents and school, which often resulted in early drug use, social withdrawal, and/or aggressive and antisocial behaviour. Several participants reported being hospitalized in psychiatric wards and/or spending time in juvenile detention centres during their childhoods and adolescence. Few participants described these early interventions as helpful; most recalled being told that they were trouble-makers and attention-seekers and internalizing a sense of being unwanted by society. Other participants reported that symptoms of mental illness “crept up on them” and gradually worsened over time as a result of accumulated stress and loss.
Many participants described not knowing how to get the help they need for their mental illness resulting in chronic, untreated illness in addition to longstanding social exclusion. A majority of participants described very negative experiences with hospitals, doctors, and medications, and were generally averse to taking psychiatric medications. Participants identified a number of gaps in the current system of care that, if in place, would help them cope with mental illness including; affordable housing, good quality psychotherapy, substance abuse counseling, and peer mentorship.
Discussion
The introduction of novel methodological combinations within qualitative research has the potential to advance inquiry into the experience of health equity for individuals as well as highlight broader structural inequities. We have presented the example of narrative interviews in conjunction with personal timelines, based on our work with individuals who experience homelessness and mental illness. We argue that the use of these methods in tandem provides a fruitful approach to examining the varied ways in which access to the SDoH influences the experience of inequity and cumulative adversity for people who are homeless.
The methods illustrated in this study contribute to the research literature in two important ways. First, blending personal and structural perspectives is particularly important in homelessness research that seeks to gain an insider perspective on the experience of inequity and systemic injustice. These integrated perspectives are needed to expand empirical and theoretical foundations of health equity. Despite not being the explicit focus of the interviews, examples of perceived inequity featured prominently in the narratives.
Second, constructing timelines in conjunction with the narrative interview facilitated the organization of rich, narrative data and allowed us to examine trajectories of events and experiences. These timelines anchor key life events as they relate to important developmental stages (e.g., early childhood, adolescence) and facilitate the examination of cumulative adversity (Kubiak, 2005). The majority of participants experienced family conflict, some form of physical or emotional abuse during childhood, and left home at an early age. For most, the onset of drug use began in early adolescence, and many participants had been hospitalized as children for mental health and/or behavioural issues. The use of timelines in conjunction with narratives allowed us to reconsider the “cause behind the cause,” by placing the construct of inequity within the context of access to the SDoH across the lifespan (Berends, 2011; Gramling & Carr, 2004). Additionally, timelines revealed the impact of various SDoH and structural inequities that were not apparent in words alone, facilitating comparisons across individuals such that a pattern of systemic inequity was revealed.
Our thematic findings of social devaluation, feeling trapped, and lack of autonomy build on previous research focused on homelessness and social inclusion, the effects of place on health, and ideologies surrounding personal choice versus structural constraints. For instance, Rowe, Kloos, Chinman, Davidson, and Boyle-Cross (2001) discuss the concept of citizenship within the context of linking homeless persons who are experiencing mental illness to their communities. Individuals experiencing homelessness and mental illness often have limited contact with mainstream society and thus fall into the category of “non-citizens,” such that they are prevented from having the opportunity to access the rights and responsibilities of the mainstream, and thus are effectively socially excluded. Awareness of a lack of citizen rights and a low position in the social hierarchy was often present in the words of the men and women we interviewed, forming the backdrop to expressions of shame, indignity, and resistance.
Restricted opportunities to access the SDoH were particularly evident in the individual and aggregated timelines. Participants in this study shared patterns of early adverse experiences, institutional exposure (e.g. foster care, hospitalization, jail), and lack of access to supportive services. Marginalization was exacerbated by long-term exposure to structural inequities, which increased participants' vulnerability to various adverse outcomes (Washington & Moxley, 2008). These findings support a complex trajectory of events and experiences that precede and follow the first episode of homelessness. In other words, homelessness is socially located, arrived at through the union of numerous individual and structural factors, which culminate in various social and health inequities (Frolich, Ross, & Richmond, 2006).
The fact that the majority of homeless shelters, support services, and resources are concentrated in a disadvantaged area of the city communicates to study participants their diminished social status. The DTES is viewed by many of our participants as “the end of the road.” There is a growing body of qualitative work that focuses on how individuals understand and experience place – the intersection of social and physical spaces (Massey, 1994). Place involves networks of social relations, and is experienced differently by different people. For people who are homeless, spatially located experiences such as neighbourhood quality are a reminder of their position in society as compared to their housed counterparts (Radley, Hodgetts, & Cullen, 2005).
Societal views of homelessness have been framed by discourse surrounding individual versus structural interpretations of public space (Cronley, 2010). For people experiencing longstanding marginalization, an apparent unhealthy behaviour may be a rational response to a constrained reality. However, such constrained choices occur within a socio-political context that promotes the individual above the collective. From this perspective, homelessness is often seen as an individual failure with limited recognition of broader structural factors. While many researchers have underscored the interplay among individual and structural factors, public views of homelessness remain largely dominated by themes of individualism, self-reliance, and choice (Caton, et al., 2005; Zinn, 2005). As such, the prevailing societal discourse is one of homeless people standing in opposition to “the public” and as a threat to the community at large (Buhler, 2009). This community at large, which presumably excludes those who are homeless, employs various forms of social control to regulate where people who are homeless can live and how they should behave (Atkinson, 2003). The conversation on homelessness needs to challenge the ways in which citizenship rights and responsibilities as a member of the public are withheld from marginalized populations.
Implications for Policy and Practice
Our findings have important implications for policy and practice relevant to this marginalized and under-served group. The early adverse experiences and ongoing accumulation of adversity are consistent with previous research showing the life course antecedents of homelessness, poor health, mental illness, and substance abuse (Shelton, Taylor, Bonner, & van den Bree, 2009; Shibusawa & Padgett, 2009; Turner & Lloyd, 2003). These antecedents highlight the importance of coordinated prevention and early intervention efforts as well as integrated services for this population at multiple points along the trajectory of homelessness.
Collective social action is required to achieve an equitable distribution of power amongst communities, institutions, and nations (Marmot, 2007). Navarro (2009) argues that the avoidance of issues of power (e.g., class, gender, race, nation state) and how power is produced and reproduced in political institutions results in apolitical recommendations that fail to address systemic inequities in health. Kenny et al. (2010) similarly critique the traditional concept of social justice, suggesting that a relational perspective should be considered such that policy processes are fair and inclusive, as well as responsive to structures of systemic inequality. In order to address the concerns of individuals, communities, and populations, public health research must make clear the varied and complex ways in which individuals are connected to groups. Qualitative methods, such as those illustrated in this paper, are uniquely capable of focusing attention on these complex interactions, and contribute knowledge and perspectives that might otherwise be absent from social policy deliberations.
Challenges and Limitations
As researchers, we are acutely aware of the “challenge of reciprocity” (Connolly, 2007) in creating meaningful relationships and findings that accurately represent interactions between research and marginalized communities. While conducting the narrative interviews, we considered our privileged roles and responsibilities with respect to the importance of telling stories that might otherwise not be heard (Adams, 2008). We also grappled with how to meaningfully represent participants' views when they were experiencing symptoms related to their mental illness. On occasion, psychotic symptoms made it difficult to follow some participants' thought processes. We decided that each individual's subjective truth should be valued equally, and included their narratives in the analysis.
Some participants may have under-reported the effects of mental illness, substance abuse, physical illness, homelessness, and their diminished status due to fears of stigma and an awareness of the power imbalance between interviewer and interviewee (Vanthuyne, 2003). We tried to minimize this power imbalance by using peer interviewers who had lived experience of homelessness with university-based researchers who had extensive experience working with this population.
Our decision to construct personal timelines developed from an effort to organize large amounts of narrative data. Therefore, timelines were constructed after the interviews (i.e., not in collaboration with participants) and this may have hindered the completeness of some timelines, as it was not always possible to locate events within corresponding developmental stages from the information provided. Ideally, it would be best to construct timelines with participants or to go back and check timelines with participants.
The aggregated timeline was a useful mechanism to represent large amounts of data, while highlighting the impact of underlying factors such as time and the SDoH. However, this form of representation has several drawbacks. An aggregate timeline emphasizes commonalities but may obfuscate the heterogeneity and differences that contribute to chronic homelessness and mental illness. Using individual timelines combined with case vignettes may create a more comprehensive, in-depth account. Also, combining typical with individual timelines could address multiple levels of analysis.
Timelines also may enforce a linear organizational framework when, in fact, most narratives are episodic rather than temporally organized. Future work could focus on the microanalysis of a few cases to build theories that relate language and meaning in ways that are missed when transparency is assumed, as in thematic analysis.
Conclusion
Our analysis demonstrates that qualitative methods, namely narrative interviews in conjunction with personal timelines, can illuminate the experience of inequity for people who are experiencing homelessness and mental illness, while revealing links between micro- and macro-level dynamics. Further innovations in qualitative methods can integrate micro and macro processes as they relate to the experience of inequity for diverse groups, and provide new knowledge that is accessible to policy makers and the general public.
Researchers armed with innovative qualitative methods have the potential to deliver new knowledge about inequity, and to highlight where opportunities for redress would have the greatest impact for improving population health.
Footnotes
Acknowledgements:
The authors would like to thank the individuals with lived experience of homelessness and mental illness who shared their stories with us; the journal reviewers for their helpful insights and suggestions; and the At Home/Chez Soi Project collaborative at both national and local levels. This research was made possible through a financial contribution from Health Canada and the Mental Health Commission of Canada. The views expressed herein solely represent the authors.
1.
Pseudonyms were used to maintain confidentiality of the study participants.