Prioritizing Prevention: Examining Shelter Diversion as an Early Intervention Approach to Respond to Youth Homelessness

Abstract

There is a growing movement in Canada towards youth homelessness prevention. One such response, called shelter diversion aims to move young people into safe and supportive housing as quickly as possible. The objective of this project is to assess how, and in what ways, shelter diversion operates and whether this intervention permanently or temporarily diverts youth from homelessness. Our project is grounded in principles of community-based participatory research including community/university partnerships and an advisory committee of lived-experience experts. Our team is utilizing mixed methods to capture the outcomes of diversion programs. Data collection began in September 2022 and ended in May 2024. Quantitative and qualitative data analysis is underway. Preliminary results show differences in how diversion is defined and implemented across organizations. There are also differences in staffing models and program budgets. The findings from this study will contribute to a recommendation for a national definition and adaptable program model for shelter diversion, easily accessible to support the expansion of diversion programs into youth-serving organizations across Canada. This study is the first in Canada to examine the effectiveness of shelter diversion as an early intervention strategy to prevent youth homelessness on a national scale.

1. Introduction

Homelessness is a persistent and problematic social and public health issue in Canada. It is estimated that an average of 235,000 people will experience homelessness each year with more than 25,000 on any given day [1]. Particularly troubling, almost half of people experience their first episode of homelessness before the age of 25 [2], with approximately 40% of those taking place before the age of 16 [3].

There is a breadth of research on the harmful impacts of homelessness including worsening physical and mental health, unsafe substance use, and increasing risk for an early death [4,5,6]. There is also growing consensus that unresolved childhood trauma is a strong predictor for homelessness [7]. Homelessness for youth is particularly problematic because young people are often fleeing very recent experiences of trauma and are at high risk for violence and exploitation once in homelessness. The stress associated with homelessness can have long term effects on physical and mental health, as well as brain development [8]. Youth experiencing homelessness often cycle in and out of institutions in addition to shelters and street life. For example, results from the 2019 National Survey of Youth Homelessness, showed that 57.8% of youth participants reported some involvement with child protective services [3] and almost 40% of homeless youth in Canada have been incarcerated [9].

There is a growing body of research that shows that youth from racialized and Indigenous communities and youth who are 2SLGBTQ+ are particularly vulnerable to homelessness and its harmful effects because of experiences of racism, discrimination, transphobia and homophobia, in addition to the issues noted above [10]. It is estimated that 2SLGBTQ+ youth comprise at least 30–40% of youth in homelessness [3]. Similarly, on any given night, one in five Indigenous people in Canada experience homelessness. This compares with 1 in 128 for the general population [11]. Many young people with intersecting identities face strain and abuse in their homes as well as in their attempts to seek support [12]. Young transgender women of colour are reportedly among the most discriminated against groups of people in shelters [10]. These issues are exacerbated by growing concern about changes to health, education and sport policies directed at transgender youth. These enhanced “parental rights” policies limit youth choices about their identities and in some cases enact criminal charges against health professionals who do not abide [13]. Advocates warn that these policies could cause significant harm including fear mongering, mistrust and panic, further increasing unsafe situations, including violence towards young people [14]. Youth who experience homelessness are also at very high risk of becoming chronically homeless into adulthood [15].

Responses to Homelessness

There are currently three primary responses to homelessness: (1) preventing it, (2) responding with crisis support like emergency shelters and (3) moving people from shelter into supported housing [3]. While cities across Canada are responding to homelessness in varying degrees, efforts to prevent homelessness are less well-developed than crisis responses and supportive housing programs [16]. To date, most of our homelessness responses in Canada support people after they have already become homeless. For example, several cities across Canada have implemented plans to end homelessness, a priority in these plans is a Housing First (HF) approach which is meant to support people in homelessness into housing with support, without requirements for sobriety or adherence to strict care plans [17]. Canada’s persistent affordable housing shortage limits the potential for the expansion of HF programs and critics argue that without a significant shift towards prevention, thousands of Canadians will continue to fall into homelessness each year [18,19].

Countries including Australia, Scotland, Ireland, Finland and Wales have embedded homelessness prevention strategies at the national level, some starting as early as the 1990s. Examples include school-based programs, human rights legislation that mandates that housing is a basic human right and collaboration within government to prevent discharging to homelessness from institutions [17].

A growing number of researchers argue that the prevention of youth homelessness, should be prioritized in service delivery and policy decisions [20] as preventing homelessness can reduce costs and system strain [17,21,22,23]. Given the complex relationship between the pathways into homelessness and the negative consequences of being in homelessness, solutions must be upstream, evidence-based and guided by the lived experiences of youth.

The purpose of this paper is to describe current approaches to youth homelessness prevention, with priority focus on shelter diversion. We also present the project protocol for a multi-city study currently underway to evaluate the impact of shelter diversion as a potential early intervention approach, meant to prevent homelessness for youth.

2. Background

2.1. What Is Homelessness Prevention?

Current public and social policies in Canada are often siloed and fragmented because of the specific mandates of government ministries (i.e., housing, or health or justice) or between the levels of government (municipal housing policy, provincial health policy or federal policy for newcomers or Indigenous peoples) [24,25]. For example, if someone is discharged from hospital, jail or a treatment centre without safe housing to go to, they are being discharged into homelessness. If a family must leave shelter because there are time limits for how long they can stay there, they may have to return to a violent situation or face homelessness. If a refugee loses their sponsorship because of a dispute or relationship breakdown, they are not eligible for most public and social benefits like affordable housing or income support. Because of these barriers, researchers have argued that homelessness prevention strategies need to be targeted at multiple levels. First, at the individual level to mitigate job loss, provide supports for mental health issues, family violence, and/or substance use. Second, at the structural level to combat poverty, racism, discrimination, colonization and the lack of affordable housing. Finally, at the system level, including gaps and siloes between government ministries and funding priorities [17]. Addressing these three levels means creating interventions for primary, secondary and tertiary homelessness prevention. Primary prevention means working far upstream including through poverty reduction strategies, ensuring an adequate supply of affordable housing and early childhood supports. Secondary homelessness prevention strategies respond to people in imminent risk including through eviction prevention, landlord tenant mediation and/or family mediation. Tertiary prevention strategies help people to exit homelessness quickly and reduce recurrence, including through Housing First programs [26].

Typically, the goal of prevention is to identify risk and protective factors and implement strategies and policies targeted at individuals or groups deemed high risk [27]. This means, there is no “one size fits all” approach to homelessness prevention. Changes are required to better align access to health care, reduce or eliminate discharge planning from public systems into homelessness, implement poverty reduction and income security strategies, and improve access to affordable housing [17,28]. A 2016 study found that robust prevention strategies could reduce public costs associated with homelessness from approximately CAD 56,000 per person per year to less than CAD 15,000 [29].

2.2. What Is Youth Homelessness Prevention?

“Youth homelessness prevention refers to policies, practices, and interventions that either (1) reduce the likelihood that a young person will experience homelessness, or (2) provides youth experiencing homelessness with the necessary supports to stabilize their housing, improve their wellbeing, connect with community, and avoid re-entry into homelessness”

[30] p.4.

One of the strongest predictors of youth homelessness is family conflict or breakdown [31]. This means that homelessness prevention for youth requires rapid housing stabilization, in-home supports to mediate conflict, transition supports for aging out of government care and discharge from institutions, low barrier and harm reduction supports, but also supports to deal with trauma, build assets, resiliency, and improve well-being [32]. This also means, and as noted above, strategies must cross public, health and social sectors. In Canada, provinces have authority over health, education, child welfare, financial income support and homelessness funding, programs and policies. Each of these is typically overseen in distinct and separate ministries who have distinct and separate mandates. For example, in Alberta, the Ministry of Community and Social Supports has a mandate to increase housing affordability but has no oversight for discharge policies from foster care or group home care [33]. The Ministry of Children and Family Services has no mandate for homelessness prevention [34] and so, in practice, as young people turn 18, they can be discharged or “age out” of financial and housing supports directly into a homeless shelter for adults. While one Ministry is working to respond to homelessness, others are creating pathways into it.

While there are many programs meant to prevent homelessness, shelter diversion has been identified as an important practice within the emergency shelter system, and some experts have argued that every community should implement shelter diversion programs [35].

2.3. What Is Shelter Diversion?

Shelter diversion is an early intervention strategy that aims to stabilize housing situations quickly, ideally with in-home supports, to prevent people from entering the shelter system in the first place, or staying there long-term [18,35]. Shelter diversion programs for youth often use a natural support approach or family re-unification strategies, which means resolving family conflict so the youth can be supported to stay at home or with the extended family. If this is not possible, finding community-based affordable housing options as quickly as possible. Shelter diversion means investing time and resources the first time a youth accesses a shelter, assessing a young person’s housing options, providing immediate supports to move them there and then supports to stay there safely. Depending on the housing they go to, supports could include financial supports, mediating family conflict and strengthening family relationships. Diversion is believed to be effective because it can reduce the number of youth that enter the shelter system, which can cut down waitlists and reduce demand for shelter beds, as well as potentially free up time and resources for shelter staff to better respond to those with more complex issues and/or no safe housing options [18].

The growing body of research on the pathways into and barriers out of homelessness, as well as research on the troubling effects of homelessness on youth safety, health and wellbeing has grown in the last two decades. This, in addition to the front-line experiences of shelter staff, led to some Canadian organizations shifting practices towards early intervention and homelessness prevention strategies, including shelter diversion. In 2018, youth shelter diversion was formalized as a pilot project led by Argus House in Cambridge, Ontario. An evaluation of their program’s results showed very promising reductions in youth stays in shelter. For example, Argus staff reported that 34% of youth were successfully diverted and 94% of those did not return to shelter. For the 6% that did return, they returned only once [36]. Following the evaluation results from Argus, between 2019 and 2021, the RAFT in St. Catherine’s/Niagara Region in Ontario, analyzed data from 379 unique youth between the ages of 16 and 24 and showed that an average of 36% were successfully diverted from shelter into housing. This number increased to 64% if the youth were diverted at their first interaction with the emergency shelter. Youth who had previous histories of homelessness showed a 27% diversion rate [37]. This data was similar to the outcomes reported in the Argus pilot study and further emphasized the importance of early intervention to divert youth from homelessness as quickly as possible. Results also showed that shelter diversion is cost effective, and relatively easy to implement. A positive diversion outcome (e.g., staying out of shelter) was dependent on assessment tools that are strength-based and include youth in decisions about housing options [37]. Based on these results, other organizations began adapting and implementing shelter diversion into their own programs for youth. Trellis Society in Calgary Alberta, Covenant House in Toronto, Ontario, Wyndham House in Guelph, Ontario, and Boys and Girls Club in Kelowna, British Columbia, were among those early adopters. These six organizations have formed a community of practice (CoP) to share learning and build awareness and understandings of shelter diversion.

Currently, diversion definitions differ across organizations, but all are typically based on how long youth stay in a shelter before they can be moved into housing. Some suggest a maximum shelter stay of 24 or 48 h, others suggest one or two weeks. Much of this is dependent on whether youth can safely be returned home with support, the number of staff in an organization working on diversion and/or the accessibility of affordable and supportive housing options in the community. Regardless, the primary goal is to support youth into safe and appropriate housing as quickly as possible and to build supports in that home to improve stability and prevent future episodes of homelessness.

The promising results noted above were the primary reason to launch a multi-city study to examine diversion practices/models across Canada. Staff from the Trellis Society of Calgary approached researchers at the University of Calgary to partner on a research project to better understand how diversion programs for youth were operating and to understand how well they were working to prevent youth homelessness. This team then reached out to the other early adopters and invited them to be partners. This larger team designed the research project to examine risk and protective factors for effective diversion and to collect stories from diversion staff, youth, and their families about the diversion experience. The results from this study will better identify whether diversion is an effective early intervention strategy to prevent youth homelessness long-term and if so, inform the development of standardized, yet adaptable programs and staffing models, budgets, training materials and toolkits. We hope to enable youth serving organizations across Canada to move towards youth homelessness prevention more broadly, including adding shelter diversion programs in their communities.

3. Materials and Methods

This study, funded by Making the Shift: Youth Homelessness Social Innovation Lab, (MtS) is a three-year research project that is currently underway with an anticipated end date of December 2024. In addition to this project, the Canadian Observatory on Homelessness (COH) and the Canadian Alliance to End Homelessness (CAEH) are national organizations who are leading or supporting research on shelter diversion for other groups including adult singles and families, so this project is part of a larger effort to more clearly understand diversion practices within multiple organizations and subgroups of the homelessness community.

Our project is applying the principles and practices of community-based participatory research (CBPR) and justice, equity, diversity and inclusion (JEDI) as our framework.

3.1. Understanding Justice, Equity, Diversity and Inclusion

Equity is the moral framework that guides the fair and non-discriminatory treatment of each individual. Not to be confused with equality, equity does not mean treating each person in the same way, but rather meeting that person’s needs while considering their unique background and situational context [38]. Diversity is the myriads of characteristics that make each person unique. These include, but are not limited to age, sex, gender, race, ethnicity, nationality, religion, sexual orientation, ability, parental status, and educational background [38]. Inclusion means embracing and celebrating difference. It is the willingness to accept every person as who they are, and to honour and respect that each individual can make meaningful contributions to their family, community, and society [39]. Justice means the valuing of all peoples and their rights, which requires action to change social conditions that privilege certain people over others [40]. Structural racism in Canada’s public systems [41] and the harms of colonization [42] mean that there are historically excluded and underrepresented populations and action regarding these unbalanced conditions is needed to provide opportunities to all groups. Youth in homelessness represent diverse groups, identities, and experiences and, therefore, our research team includes members with diverse and intersecting identities and their own lived experiences. Our team is also transdisciplinary, including members from academic disciplines such as disability studies, critical social work and feminist and race studies. Each research team member, regardless of age or research experience, has an equal voice and opportunities to contribute, challenge and debate. Our community-based approach creates a balance between academic and frontline experience which ensures that our findings and subsequent recommendations will be guided by and grounded in rigorous methodologies and “real life”. As such, JEDI is both a strategy and an outcome of our project.

3.2. Community-Based Participatory Research

Community-based participatory research (CBPR) is an approach meant to reduce gaps between evidence-based research results and practice and policy change. CBPR transforms the research relationship and prioritizes social action as the primary purpose of research [43]. Collaborative community and academic research require a significant investment of time and resources. When research does not translate into changes to policies and programs, it fuels an ongoing degree of mistrust, frustration, and unwillingness of decision makers to participate in future partnerships [44]. It is necessary for our project to include lived-experience experts and service providers within the research design to inform evidence-based changes to policy and service delivery. Political inaction results from “ivory tower” research that is detached from the realities of daily life for the purposes of publication in academic journals that are only accessible to other academics. Research that is carried out solely for the purpose of academic publications, that only reach other academics, prevents changes to policies and practices and creates the same types of gaps as siloed government ministries and mandates.

An advisory committee of lived-experience experts is actively engaged and helped develop consent forms, ethics processes and data collection tools and will support the analysis of results and any implications, deliverables and recommendations for policy and practice that emerge. The Youth Advisory Committee (YAC) meets monthly and includes between four and six young people and a committee chair. The YAC members represent three provinces across Canada and are paid members of the research team. Other members of the research team attend YAC meetings as invited guests and notes are taken which are being used continuously to inform adaptations to the study. We are currently co-developing a manuscript based on the ethics of research and practice from the perspectives of lived expertise, including how to develop trust on the research team, avoid exploitation and the use and consequences of inclusive/exclusive language.

3.3. Mixed Methods

We are utilizing several sources of data including agency intake data and interviews in an ethnographic case study [45] and have included a developmental evaluation to ensure we are capturing the outcomes of diversion programs (what happened) as well as the process involved in implementing diversion programs (how it happened). This project prioritizes the lived experience and expertise of staff working in diversion programs and youth and families who are or have accessed diversion services.

Case study research is appropriate, as we are interested in examining the background, context and processes associated with diversion as well as experiences of participants and staff [46]. Our developmental evaluation allows for a deep understanding of the nuances and particulars of diversion programming in the organizations in our study and how they are different or similar to each other [47]. Our intention is to understand how, and in what ways, shelter diversion operates and whether this intervention permanently or temporarily diverts youth from homelessness. In so doing, we can assess alignment, strengths, and accomplishments across organizations as well as areas for growth and development.

3.4. Research Questions

The research questions guiding this study are organized into three areas: helpful components of the diversion model, housing outcomes for youth and families, and suggestions for program improvement. The primary research question guiding this study is the following: Does shelter diversion prevent the experience of homelessness or merely delay it?

Secondary questions include the following:

• How does the diversion program work? What are the similarities and differences in models across the sites?
• What are the perspectives of young people (and their families) on the effectiveness of the intervention?
• Are there important differences between large and smaller municipalities (Toronto, 2.8 million people; Calgary 1.3 million; Peterborough 85,000; St Catherines/Niagara Region 420,000; Kelowna, BC, 156,000; Guelph, Ontario 152,000, Cambridge Ontario, 150,000).
• What do the results tell us about how to further improve programs and diversion outcomes for youth with diverse and intersecting identities?

3.5. Recruitment

Youth and family member participants are being recruited through the partner sites. Shelter staff let potential participants know about the study and provide contact information for the research team. We utilized purposive sampling (recruiting participants with the characteristics we need for the study), as we are interested in learning about the experiences of youth and families who have contacted a shelter and engaged in shelter diversion services. All participants provided signed consent to participate. Our ethics approval rules allow for young people aged 14 to 18 to provide their own consent, if after an initial meeting with a researcher, they were confident that the youth clearly understood the purpose of the study and they had the choice to withdraw or end the interview.

3.6. Data Collection

This is a three-year study; however, year one (2021) was spent developing relationships with each site, recruiting the YAC and obtaining ethic approvals. We collected several types of, and sources of data between summer 2022 and summer 2024. Although case studies do not typically utilize survey data, we are analyzing administrative data (youth shelter intake records) to examine participant demographics and percentages of youth who were diverted from shelter. Each shelter collects different data; however, common data points include age and gender identity, reasons for accessing shelter, length of shelter stays, site of housing placement, referrals and return to shelter (or not). This data will help us better understand “who” is accessing shelter services, how many and under what circumstances, as well potentially, if once supported into housing, how many young people returned to shelter. Each site is collecting its own data and de-identified data has been “rolled-up” for comparative analysis, where possible. Results will primarily be presented as descriptive statistics with some potential for correlation and causal analysis if the sample sizes are large enough. For example, the percentage of participants who identify as male, female, transgender or non-binary and their length of stay in shelter. It is important to note that not all of the sites collect data on race and so we will not, in this study, make a separate analysis of race.

Qualitative data includes individual interviews with youth and/or family members accessing diversion programs within the six sites. The purpose of these interviews is to capture experiences with diversion, including interactions with staff, services accessed, and helpful and not helpful aspects as well as hopes and dreams for the future. Individual and group interviews with staff examine the delivery and management of the program, aspects that are “working or not” and suggestions for further improvement.

Data collected for the DE include internal program documents, program definitions, and staffing models as well as interviews with initiators of the program to understand its history and development. We are also using research team and advisory committee meeting minutes to examine our own processes and decision making.

Our data collection strategy includes the following:

• Review of previous evaluation reports, organizational documents regarding program models and definitions from the partner sites.
• Some 400–500 unique youth records taken between 2022 and 2024 from intake data collected in the diversion programs at the partner sites.
• Interviews with ~50 front-line workers and managers in diversion programs.
• Approximately 100 interviews with youth and family members who have accessed diversion services with one or two follow-up interviews where possible.
• Quarterly research team meeting notes.
• Monthly team meeting notes from the Youth Advisory Committee.

3.7. Data Analysis

Quantitative data will be largely descriptive, SPSS is being used to analyze organizational intake data. The qualitative interview data will be analyzed through a six-step thematic coding process from Braun and Clarke that includes data familiarization, code generation, theme construction, theme review, theme definitions, and knowledge translation/dissemination strategies [48]. The research team includes six research assistants who work in pairs to code the data then the larger research team meets to discuss the codes and create the themes.

4. Objectives and Anticipated Outcomes

The overall objective of this project is to understand if diversion programs can prevent homelessness and if so, to develop adaptable shelter diversion program models, for youth at imminent risk of homelessness and their families. While we do not expect diversion to be a panacea, or an intervention that is appropriate for everyone and in every case, we anticipate being able to influence other communities in their exploration and implementation of diversion as an early intervention program. We recognize the necessity for a full continuum of prevention strategies targeted at individual, systems and structural issues as well as primary, secondary and tertiary prevention.

We anticipate creating a definition of shelter diversion with recommended data points for intake as well as for research and program evaluation. We anticipate being able to highlight particular subgroups or risk factors that could require additional funding, partnerships and/or services (e.g., youth with diverse and intersecting identities and/or youth who have past histories of homelessness who cannot rapidly be supported back home). We also anticipate recommendations for changes to government policy to reduce some of the structural barriers many young people face, including siloed public systems, a lack of discharge planning and aging out of care (for example). We anticipate several publications of research findings as well as the processes and CBPR and JEDI principles and practices we are following. Community forums will occur in each city to share findings locally and a national webinar(s) will share findings more broadly. Finally, we plan to collaborate further with national organizations like MtS, COH and CAEH to support a comprehensive and inclusive “diversion” movement based on our results.

5. Discussions

Strengths and Limitations

The strengths of this study lie in the community-based participatory approach, and the ambitious data-collection strategy. This is the first multi-city study on shelter diversion to be carried out in full partnership between lived-experience experts, universities and community service providers. Another strength is the focus on mobilizing the findings into tangible strategies to support the scale up of a novel early intervention program meant to prevent youth homelessness. Limitations include the variance in how organizations are defining shelter diversion and the differences in the data being collected from each organization. Program models vary across organizations including, staffing models and budgets and the interventions being offered in diversion programs including family mediation and reunification efforts. Each community has access to different funding sources, differences in political will, community resources and affordable housing units. Our study may identify particular subgroups of youth who are not able to be diverted safely; however, specific strategies to address racism, discrimination, trans and homophobia may be outside the realm of this study. Future research should build on the foundational learnings of this project to more rigorously examine safe practices for youth with multiple and intersecting identities.

6. Conclusions

Homelessness prevention strategies are often nebulous and less developed than crisis responses; however, they are an essential part of the continuum of supports needed for youth living in or at risk of homelessness. While several organizations in Canada are shifting towards homelessness prevention, as far as we know, this is the first multi-city study that is examining a potential early “best-practice” intervention called shelter diversion, meant to keep youth out of shelter and support family reunification through a natural supports approach. By applying a community-based multi-methods approach and elevating the experiences of experts in youth homelessness namely, youth, families, and front-line workers, we will enhance and expand current shelter diversion models and assess the potential for scale up in communities across the country. Given the very promising results reported by shelter diversion projects so far, we anticipate being able to articulate clear and specific diversion practices with practical tools and recommendations for program implementation. Results from this study have the potential to influence the uptake of nationwide shelter diversion programs. If implemented as part of a full continuum of prevention strategies and if supported by changes to government funding and inter-ministerial collaboration, this program model could prevent homelessness for thousands of young people in Canada.