The Work of Youth Homelessness Prevention in Ontario: Points of Frustration, Points of Potential

: Despite a rhetorical turn towards prevention in homelessness policy and research, the work of youth homelessness prevention continues to be frustrated by persistent structural barriers. In this article, we examine how youth homelessness prevention is being implemented in the province of Ontario, with a focus on targeted provincial support programs and local shelter diversion practices. Drawing on interviews with workers in the homeless-serving sector, we describe the implementation of these initiatives and identify points of frustration and potential that workers encounter as they try to prevent experiences of homelessness for youth. We contend that these points of frustration illuminate persistent structural barriers that continue to forestall the work of youth homelessness prevention. Meanwhile, points of potential demonstrate the importance of empowering workers to creatively adapt and offer responsive services. Taken together, these signal the critical importance of two aspects of contemporary homelessness prevention typologies—primordial prevention and empowerment. We end by offering aspirations for action, a political reframing of the policy recommendations sections more typical of social science research articles. We do so to affirm our commitment to advancing the work of structural transformation that is required to achieve the right to housing for all, including youth.


Introduction
Experiences of homelessness are associated with several other challenges for youth.These include poor mental [1] and physical health [2]; exposure to violence [3]; high levels of educational disengagement [4]; challenges associated with substance use [5]; and criminalization [6].A young person's experience of homelessness has the potential to disrupt important developmental processes, which can negatively impact their future physical and emotional health outcomes [7].Furthermore, young people who identify as 2SLGBTQIA+ as well as Indigenous and racialized youth are over-represented among those experiencing homelessness in Canada [8].
In light of these associations, scholars across the Global North have increasingly come to conceptualize youth homelessness as a population health issue with significant and potentially long-term repercussions for health equity.Since the 1990s, frameworks for disease prevention that were first developed in the field of public health have increasingly been adapted and applied to this social issue [9][10][11][12], and now calls for "homelessness prevention" are mounting.
In this article, we examine two approaches to youth homelessness prevention currently being pursued in the province of Ontario: the provincial Youth in Transition Worker (YITW) and Housing Support Worker (HSW) programs, and local shelter diversion initiatives.We focus on data practices employed by people working in these programs to understand how public sector modernization efforts are playing out at the program level.The Youth in Transition Worker (YITW) and Housing Support Worker (HSW) programs are provincial initiatives designed to provide targeted support to youth involved in the child welfare system or otherwise in need of protection.While there is currently no provincial directive for shelter diversion in Ontario, it is identified as one of five eligible activity streams within the federal Reaching Home strategy [13] and is being implemented locally by municipalities and service agencies across the province.
Drawing on interviews with frontline workers and program managers in the child welfare and youth homelessness sectors, we identify points of frustration and possibility experienced by workers as they implement these programs.In doing so, we illuminate some of the tensions that arise in translating youth homelessness prevention policy recommendations and program models into frontline work environments.Particular attention is paid to how these prevention initiatives fit within the wider context of homeless-service provision in Ontario, including mandated data practices and service models.Examining these tensions brings to light the structural preconditions that are necessary for the operation of individual-level youth homelessness prevention efforts, and the importance of maintaining flexibility and responsive capacity in the delivery of frontline youth homelessness prevention services.
This article proceeds in five parts.We begin by outlining our research approach (Materials and Methods), including our conceptual grounding and processes of data collection and analysis.We then turn to a review of the literature on homelessness prevention, focusing on those frameworks and typologies that are specific to youth and have been developed and deployed in Canada (Literature Review).Next, drawing on interview data and document analysis, we dive into the empirical findings of our research to describe the delivery of the YITW and HSW programs and day-to-day shelter diversion practices (Results).Patterns of experience in these programs and practices are illuminated and used as the basis upon which points of frustration and possibility are identified.Finally, we discuss how the patterns evident in workers' experiences of implementing these youth homelessness prevention efforts can help contextualize and enliven existing prevention frameworks and typologies such that new opportunities for transformative action may emerge (Discussion).We close with a poem, developed from our interview transcripts, that offers a collective vision for supporting youth to realize their right to housing (Conclusion).

Materials and Methods
The findings of this paper draw from a set of inter-related projects being conducted at the Research for Social Change Lab based out of Trent University.The first of these projects sought to identify and map the processes by which a young person's experience in Ontario's child welfare system is systematically recorded and produced as data and how that data in turn inform service provision and decision-making.A subsequent project extended the scope of the research to the housing and homelessness system in Ontario.Applying parallel methodologies, our research teams-data justice for youth in care and data justice for youth experiencing homelessness-compiled a portrait of how data-driven and digital information management practices are operating in two interconnected social service systems in the province of Ontario: child welfare and youth homelessness services.
The primary objectives of this work were to (1) identify what data are being produced in these systems, (2) assess their quality, (3) determine the purposes for which they are being collected, and (4) describe how they are being used and to what effects.Our goal in undertaking this data infrastructure audit was to enable transparency, improve navigability, and increase capacity to hold institutions accountable for youth wellbeing.To realize these objectives, we created visualizations (or maps) of data and data infrastructure, coupled with detailed descriptions of day-to-day work practices undertaken in the provision of child welfare and youth homelessness services, including intake procedures, assessment processes, and reporting activities.These descriptions, specifically those related to the implementation of the YITW and HSW programs and the work of shelter diversion, form the empirical basis of the analysis presented in this article.
Our research approach is informed by an alternative mode of sociology known as institutional ethnography (IE).Developed by Dorothy Smith in the 1980s, IE takes as its starting point some part of the everyday world and people's activities in it.The research process involves engaging people in conversations and observing their practices to "reveal troubles arising in (or conflicts between) authorized and experiential knowledge" [14] (p.3).Following this mode of inquiry, researchers seek to illuminate "how it is" that people's experiential knowledge diverges from authorized accounts [15] (p.692), attending to the processes through which diverse people's activities are coordinated across local settings [16].Particular attention is paid to the ways in which these processes may be at odds with the interests of people in the local setting [17].While an IE inquiry proceeds from the everyday world-providing empirical descriptions of its actualities-it extends analytical attention towards the organization of social relations not immediately evident within it.
Applying this mode of inquiry to the projects at hand, we anchor our research in the everyday activities that comprise the data practices through which child welfare and homelessness services are organized.From this starting point, we have identified a disjuncture between authorized accounts of youth homelessness prevention and workers' everyday experiences of supporting youth to obtain and retain housing.In other words, we have identified a gap between the prevention turn expressed at a discursive level-circulated in policy texts and operationalized through specific program models and funding streams-and the on-the-ground realities that workers face in providing youth with the material and social resources necessary for them to secure and maintain stable housing.
As outlined above, our research has proceeded in several phases running sequentially in the child welfare and youth homelessness systems.In Phase One, we completed keyinformant interviews and desk research to map the institutional data assets of the child welfare system in Ontario.Using the map as a guide, we then facilitated focus groups with policy leaders, institutional decision-makers, researchers, former service users, advocates, and service providers to gather feedback regarding the accuracy and utility of the map we produced.After the focus groups, we applied what we learned to improve the map and produce a new version.We then replicated this process in the youth homelessness system (Phase Two).In Phase Three, we conducted interviews with homelessness service providers and child welfare workers (of which only two had participated in early focus groups) to dig deeper into some of the issues raised during the first two phases of the project.Interviews were conducted online using a co-interview protocol and focused on illuminating workers' data practices and the ways these are shaped by digital infrastructure (e.g., data fields), tools (e.g., assessment protocols), and policy mandates (e.g., homelessness enumeration).
In total, we interviewed 72 unique participants for this study.The analysis offered here is based on a subset of people who worked in or directly with YITW, HSW, or diversion programs: 24 unique participants in 21 individual interviews, one software demonstration, and 1 group interview conducted with 3 individuals who had previous experience working at the same agency.Of these individuals, fifteen had experience with shelter diversion (as frontline workers, or managers of homeless-serving systems in which shelter diversion was being undertaken), and five were working, or had worked, in YITW or HSW positions across south-central Ontario.
In addition to interviews and focus groups, our research included text and document analysis that began during Phase One and continued throughout data collection.Employing Smith's broad conceptualization of texts-that is, replicable artefacts that coordinate people's practices across time and space [16]-when references to an organizational or policy text (for example) were made by our participants, we looked them up online or asked for copies to review.The combined use of textual and interview analyses has allowed us to identify how patterns of experience in the provision of youth homelessness prevention services are organized by social relations originating elsewhere and taken up through people's engagement with texts.
In institutional ethnography, the identification of common experiences-patterns in how people talk about their work-is the first step in analysis.From here, researchers seek to explicate the textually mediated relations shaping the patterns that have been identified [16].The intention is not to generalize people's experiences as instances of a theme, but rather it is to explain how the experiences people describe are organized by generalized institutional relations.Our data processing and analytic practices advance this aim.Interview transcripts were indexed to identify specific types of data (e.g., identification), data collection instruments and assessment tools (e.g., the Service Prioritization Decision Assistance Tool, SPDAT), data infrastructure (e.g., HIFIS), and data practices (e.g., intake).We also indexed references to specific policies, legislation and funding programs (e.g., the Housing Services Act, Reaching Home) as well as utterances of professional jargon and discourse (e.g., housing first).Finally, we coded our interview transcripts for two overarching themes-data aims (things people are seeking to do with data) and data desires (things people wish they could do, but currently cannot).This process of bringing into focus the institutional contexts shaping people's work is emblematic of institutional ethnographic ways of working with interview data [18].

Literature Review
In Canada, the emergence of homelessness as a widespread "social problem" can be traced back to the 1980s and 1990s [19].It was at this time that "individuals, service providers, faith groups, the non-profit sector and local governments worked to develop a range of services that responded to the immediate needs of people who are homeless" [19] (p.23).Now, some forty years later, calls for action have shifted from the provision of emergency services to meet immediate needs to strategies and policy directives aimed at preventing homelessness from occurring in the first place.
In Canada, the Canadian Observatory on Homelessness (COH) has led the development of different tools and resources to support efforts for homelessness prevention, including homelessness prevention for youth.In 2017, they published 'A New Direction: A Framework for Homelessness Prevention' [20].This text, in addition to setting forth a definition and case for prevention, presents a typology of homelessness prevention "intended to identify and organize the range of activities that are considered to be central to homelessness prevention" [20] (p.43).The typology is organized into five categories: (1) structural prevention, (2) systems prevention, (3) early intervention, (4) eviction prevention, and (5) housing stability [20].These five categories are used to classify types of activities that are understood to support homelessness prevention.For example, "systems prevention" is defined as activities aimed at "addressing institutional and systems failures that either indirectly or directly contribute to the risk of homelessness" [20] (p.44).
This particular typology is further framed around the public health concepts of primary, secondary, and tertiary prevention [20].Normed for disease prevention, these concepts signal when in the progression of illness an intervention is pursued.As described by Culhane [21]: Primary prevention initiatives are those which prevent new cases; where efforts focus on reducing the risk for acquiring a particular condition.Secondary prevention identifies and addresses a condition at its earliest stages.Thus, it does not reduce the number of new cases, but rather treats conditions close to their onset while they are presumably easier to counteract.Finally, tertiary prevention seeks to slow the progression or mitigate the effects of a particular condition once it has become established.(p.297) In addition to offering a conceptual organization for what kinds of intervention to deliver when, the framework identifies the population groups (who) interventions are aimed at, using the language of universal, selected, and indicated prevention approaches [20].In this schema, universal prevention approaches are those that target the entire population, where selected approaches are aimed at people who are part of a group identified to be at risk of homelessness (e.g., youth in care; people experiencing poverty), and indicated approaches target those who are at risk of homelessness due to individual characteristics (e.g., people with addictions).
In 2018, the CoH released a subsequent 'Roadmap for the Prevention of Youth Homelessness' which builds on their established typology, adding "an additional legislative strategy" known as the duty to assist [22] (p.24).As described in this text, the duty to assist enshrines the "legal duty to ensure that young people are provided with information, advice, and housing-led supports to avoid an experience of homelessness, or to make that experience as brief as possible" [22] (p.26).Thus, the roadmap outlines six categories of homelessness prevention interventions: (1) structural prevention, (2) systems prevention, (3) early intervention, (4) eviction prevention, (5) housing stabilization, and ( 6) the duty to assist.
Based on a systematic review of evidence, a further update to the original typology was put forward in 2019.This version retains the five categories from the 2017 text (now referred to as domains) and adds a sixth domain of empowerment, "consistent with the principle of nothing about us without us" [23] (p.1760).Additionally, this version of the typology updates the conception of primary, secondary, and tertiary prevention, adding primordial and quaternary prevention.Drawing on more the recent public health literature, the authors define primordial prevention as "a focus on systems and policy level preconditions of a health or social issue" and quaternary prevention as "a focus on mitigating the unintended consequences of interventions across the other four levels of prevention" [23] (p.1755).
Between 2017 and 2019, three homelessness prevention typologies were put forth by researchers affiliated with the COH: one, a generalized typology for homelessness prevention as a whole; another, a roadmap for youth homelessness prevention specifically; and the last, an update and expansion to the original framework.In the next section, we use the language and categories of these homelessness prevention typologies to unpack the work of youth homelessness prevention as it is being undertaken in Ontario today.

Results
There are a variety of youth homelessness prevention interventions operating across the province of Ontario today.Two in particular are of interest to us in this article: the Ministry of Children, Community, and Social Services (MCCSS)-funded Youth in Transition Worker (YITW) and Housing Support Worker (HSW) programs, and frontline youth shelter diversion practices.These initiatives differ in that the first is a set of provincially mandated programs of support, and the second a diffuse constellation of practices implemented within local homeless-serving organizations and systems.We selected them for analysis here because they emerged as significant nodes of activity and interest over the course of our interviews.In examining the data systems at work in the youth homeless-serving system, we were directed to speak with individuals working in YITW and HSW roles, and in our discussions with these and other frontline homelessness workers and systems managers, the concept and practices of youth shelter diversion were mentioned frequently-despite not being specifically prompted in our interview script.Below, we provide a brief overview of each of these initiatives in turn.
There are two types of provincial youth support workers tasked with improving the outcomes and increasing the resilience of youth involved in Ontario's child welfare system as they move toward adulthood: 1. Youth in Transition Workers (YITWs), focused on supporting youth leaving the care of local Children's Aid Societies (CASs), and 2. Housing Support Workers (HSWs), focused on supporting youth who have, or are seeking a Voluntary Youth Service Agreement (VYSA) (VYSAs provide a mechanism for youth aged 16 and 17 in need of protection to voluntarily apply for support through their local CAS) to find and retain housing [24].The positions share the objective of helping youth identify, access, and navigate adult service systems during the transition to adulthood.However, the target populations for the two positions differ, with YITWs providing services to young people aged 16-24 who have been in extended society care (formerly referred to as Crown Wards) or have had an active agreement with CAS, and HSWs working with youth aged 16-17 who have sought or are seeking support to initiate a VYSA.YITWs and HSWs are employed by community-based agencies that receive transfer payments from MCCSS to deliver the YITW and HSW programs.
As MCCSS-funded programs, both the YITW and HSW roles are clearly defined by the province and accountable to provincial service objectives and reporting requirements.The roles themselves are held at different community-based agencies (including but not limited to shelters) and the scope of work, though oriented towards housing and homelessness prevention, includes an array of other supports as needed on a case-by-case basis.As such, there is an expectation that YITWs and HSWs coordinate their work with those working elsewhere in the homeless-serving system, the child welfare system, and other relevant systems (such as social assistance or developmental services), while simultaneously responding to young people's evolving life circumstances.Within the dynamics of these multi-system and person-centered roles, the YITWs and HSWs we spoke to see homelessness prevention as central to their work.
In contrast to the provincial YITW and HSW programs, shelter diversion is not a provincially funded program; rather, it is identified as one of five eligible streams of activities/expenses for federal homelessness funding through the Reaching Home program [13].In that context, shelter diversion is defined as: a tool used to prevent the use of emergency shelters by providing individualized supports when families and individuals are seeking to enter the emergency shelter system.Shelter diversion programs help individuals and families seeking shelter to explore safe and appropriate alternate housing arrangements and, if necessary, connect them with services and financial assistance to help them find secure housing [13].(para.24) As described in Reaching Home program materials, shelter diversion is a strategy specifically aimed at preventing the use of emergency shelters (rather than preventing homelessness).Unlike the YITW and HSW programs, there are no established service objectives, outcomes reporting requirements, or implementation guidelines.Rather, individual agencies and municipal service managers dictate the terms and processes by which shelter diversion occurs, sometimes through dedicated program pathways or outreach services.While shelter diversion is always premised on the goal of diverting people from emergency shelters, in our interviews it was variously described as an established program, a loose set of practices, and as an overarching ethos within shelter organizations.
Shelter diversion efforts are not subject to the same type of textually mediated oversight as is evident in the provincial prevention initiatives.However, there are similarities across locally developed initiatives that point towards shared resources and tools (e.g., the use of an oral script to explore safe and appropriate alternatives when a youth presents at a shelter; and the idea of progressive engagement).Examples of such shared resources and tools include webinar trainings offered by the Canadian Shelter Transformation Network and Built for Zero Canada [25,26].In contrast to the provincially governed YITW and HSW programs, shelter diversion is a contested space, with several entities vying for influence and shaping key differences in how it is understood and implemented.
Here, we have two examples of how the work of homelessness prevention is currently being implemented in the province of Ontario: one, a provincially mandated program of supports to a group of youth determined to be at risk of homelessness, which could be identified within existing homelessness prevention typologies as a primary, selected, systems prevention initiative; the other, a loose set of practices and principles enacted locally by individual agencies or service provider collaboratives, which could be given the label of a secondary, indicated, early intervention strategy.
As we will illuminate in the following section, the ways these practices and processes work alongside mandated data practices (e.g., By-Name Lists) and service models (e.g., Coordinated Access systems) in the homeless-serving system has implications for how the work of youth homelessness prevention is, and can be, actualized.It is in this context that points of frustration and potential emerge in the work of youth homelessness prevention in Ontario.

Youth Homelessness Prevention and Mandated Data Practices
In Ontario, municipal housing and homelessness service managers are required to adhere to provincially mandated homelessness enumeration practices, including keeping up-to-date lists of all people known to be experiencing homelessness in their regions, known as By-Name Lists (BNLs).Provincial guidelines define homelessness "as a situation in which an individual or family is without stable, permanent, appropriate housing, or the immediate prospect, means and ability of acquiring it" [27] (para.32).This definition encompasses situations in which individuals are provisionally housed with friends or family (e.g., couch surfing)-the very types of temporary situations workers describe facilitating during a shelter diversion.These lists are stored in databases known as Homelessness Management Information Systems (HMISs), with the Homeless Individuals and Families Information System (HIFIS) being the HMIS supported by the federal government.
In regions receiving federal homelessness funding, service providers are also required to implement a Coordinated Access approach to triage and prioritize housing placements across their local service delivery system.As per the federal Reaching Home Coordinated Access guide, these systems make use of BNLs to generate Coordinated Access lists, which can then be used to prioritize and match people experiencing homelessness to available housing resources [28].While By-Name Lists are intended to include all (consenting) individuals experiencing homelessness in a given service-delivery region, Coordinated Access lists include only "clients who are eligible and interested in housing resources" [28] (p.15).A further refinement of this list produces a priority list which includes "clients who are able to accept an offer of housing resources immediately" [28] (p.15).
One municipal homelessness service manager described these lists and the processes they use for populating and prioritizing them this way: So, the BNL as a whole is everyone known to be experiencing homelessness.From that we create a By-Name Priority List for each men, women, youth, and families. . .And so, from the BNL, we use the criteria that we've come up with as a community for prioritization. . .right now, our prioritization is largely around chronic homelessness but it's because we're really only offering those specific types of resources through it.(Valerie) The same manager, Valerie, described being "very concerned about creating a pipeline to chronic homelessness" through the BNL and BNPL prioritization processes, noting the need for a "full continuum" of supports to ensure this does not occur: The reason we still feel comfortable with that [prioritization]-because we are also very concerned about creating a pipeline to chronic homelessness if there aren't supports earlier-it's because. . .it's more likely I guess, that somebody would need two years of case management support if they've been experiencing chronic homelessness.Someone who's new to homelessness might not necessarily need that.You kind of want to prioritize that resource for someone who's kind of stuck within the system.But that's where we really need to make sure that we have that full continuum, so that we do have the early intervention supports like prevention, first hand diversion, and then where diversion fails, really having that early intervention.So, we are also reinvesting in our shelter early intervention programs to do that sort of quick, intensive case management with people to end those early experiences of homelessness too.
Here we see that while homelessness, as defined provincially for the purposes of the BNL, includes a range of experiences of housing precarity (including couch surfing), the processes of prioritization and matching for which the BNL is foundational orients service providers towards resolving more entrenched experiences of homelessness because of the resources being allocated through it.Other necessary resources (such as "quick intensive case management to end those early experiences of homelessness") are needed to ensure the current response is not entrenching chronic homelessness.Research on critical time interventions suggests that structured, time-limited in situ (or community-based) case management improves housing stabilization for people with mental health and/or substance use challenges who are transitioning from an emergency shelter environment [29], but to date, the majority of resources accessed through the BNPL in Valerie's municipality are directed towards addressing the needs of clients experiencing chronic homelessness.
As noted above, under the federal Reaching Home program, communities are encouraged to develop their own prioritization criteria and matching processes as part of the Coordinated Access workflow (i.e., access; triage and assessment; prioritization; matching and referral).However, they are required to use a common assessment tool to evaluate client acuity.One such tool used across service delivery settings is the Service Prioritization Decision Assistance Tool (SPDAT) or the shorter Vulnerability Index-Service Prioritization Decision Assistance Tool (VI-SPDAT)-a tool which has been shown to perpetuate racial bias in housing triage decisions [30].According to the workflow described in the Reaching Home Coordinated Access Guide, shelter diversion is one of two possible processes initiated during the triage stage of service delivery, after a person at risk of homelessness first comes into contact with a homelessness service provider (the other being eviction prevention) [28].In cases where there is successful resolution to the housing challenge through access to informal or natural supports, "no next steps" are identified for the client [28] and they are not added to the By-Name List.As we heard from one diversion worker, this creates gaps in access to resources for youth who are engaging in diversion protocols: Yeah, so we do use one [SPDAT] once they're in shelter or if they're already experiencing homelessness.And I also do the prevention SPDAT, the shorter one with youth.The one thing about the prevention SPDAT is that they're not a SPDAT that you can put into HIFIS through the [municipality] and they don't kind of reflect on the By-Name List as well, which is kind of a pet peeve of mine.Because to get on the By-Name List, you already have to be experiencing homelessness.But we're seeing a lot of folks that, you know, they're getting deadlines of, 'oh, the landlords selling the house' or 'I'm being evicted at this time' or 'I'm a student and my student housing ends', you know, on January first or whatever.And we can't get them onto the By-Name list until they've already experienced homelessness, which is a huge pet peeve of mine because I feel like we can be preventing and diverting folks from homelessness if there is a process of getting them onto that By-Name List and making them eligible for housing before they're even losing their housing.(Jackie) Here we see that the data practices that are set up by provincial and federal homelessness funders may obscure, and at worst obstruct, attempts at early intervention into experiences of homelessness.While eviction prevention and shelter diversion remain named as activities to be undertaken within the homeless-serving system, these activities do not easily fit within the data processes and practices set up to account for, manage, and distribute housing resources.
While changes to these data practices could better integrate diversion within mandated data practices, ethical concerns around the collection and storage of personal information, especially for youth, would persist.Reflecting these ethical concerns, one coordinator of a youth shelter diversion program described their attempts to keep youth out of the [HMIS] system to us this way: Currently, if the youth's situation is resolved within those 14 days, they're not added onto the By-Name List, because we're not recording any information in HIFIS.Unless they pop back up again and then it's a bit of a different situation.But I'm trying to keep folks out of systems that they don't necessarily need to be involved in if that light-touch support is all that is needed.(Andrea) Here, the concern expressed hinges on a young person being recorded in an HMIS and having the tag of homelessness follow them around.This same worker elaborated on this concern and their ongoing negotiations with their municipal partner on the use of HIFIS in their program this way: My hesitation is. . .as soon as the city's like, you have access to HIFIS now and this is what we're requiring to be recorded within the program.We're really hoping that will be a conversation, because for our youth to have their data stored and the tag of homelessness follow them around, we don't want that if that's not necessary.And so, I think those conversations will still be forthcoming in a lot of ways.I hope that they'll be forthcoming in a lot of ways and that they will be conversations in and of themselves and not just, 'okay, you have HIFIS, this is what needs to be put on there'.And 'we need every youth that flows through the program to be tracked', and all that.I think that's part of why we're not on it yet.It's because it's complicated.And so, one day, we'll sort through the complications, maybe, and get there.(Andrea) In other interviews, shelter workers described more general guidelines and best practices around not over-collecting sensitive information.As one shelter manager described: As much as possible we try and kind of warn our staff against kind of collecting information that's not pertinent for anything that they can immediately use. . .all of this is sort of, you know, explained to the clients before and it's up to them to share, whether they want to or not. . .Like if we're helping them with immigration, things related to that, that might actually involve some sensitive information surrounding the basis of refugee claim, for instance.Again, things like that, there are ways for a client to kind of just work with their lawyer one-onone without support from us.For the most part, they do seek our support, which is why we kind of do get access to some sensitive information.But yeah, with the exception of the main three things that we would probably seek to obtain [consent, ID, income verification] everything else is kind of, you know, case dependent, and usually if the client wants to share.(Mustafa) Here, Mustafa shares how he supports staff to think critically about how they will use data to support a young person's life objectives, so as not to default towards the overcollection of data that has no direct utility for case work.The sensitivity of information being collected from youth at a particularly vulnerable moment in their lives (experiences of housing loss, precarity, and homelessness) is something that frontline workers and service coordinators consider as part of their everyday work practices.
This everyday practical ethic lies in tension with mandated data practices in as much as these processes rely on widening the scope and frequency of personal information collection in order to operate standardized matching and referral processes, and assess service needs.While those working to manage homeless systems are aware of the need to provide a full continuum of supports in order to avoid "creating a pipeline to chronic homelessness", current homelessness management practices like those associated with prioritization of people on the BNL continue to be geared towards the allocation of more intensive supports to those experiencing chronic homelessness.Thus, current mandated data practices appear ill suited to support the work of youth homelessness prevention, especially in the context of scarce housing resources.

Youth Homelessness Prevention and Housing Resource Scarcity
The current environment of scarcity within available housing resources (expressed as a lack of affordable housing options) was brought up frequently by our interview participants as a persistent structural barrier impeding their ability to truly prevent youth homelessness.As one YITW put it: Right now, in [municipality C], we're looking at room rentals that are anywhere between $800, $900, $1000 for a room rental.And if you're a youth who's transi-tioning out of CAS and your next step is-because you don't have employment or you haven't learned those skills to obtain employment-you're going on Ontario Works.When you only get seven something for Ontario Works, how the heck are you supposed to afford $800 room rental?It's impossible.(Mike) Mike goes on to describe how even youth with financial resources face discrimination from landlords when trying to access the private rental market: I would say the majority-at least 75-80% of my referrals have some type of support around housing, and to be completely honest, there's just not a lot of availability even for the people that have financial backing.People that are working full time-just the cost of living and landlords are so [prohibitive]-landlords, they discriminate against this population [of youth].When I first started in this field, years ago I worked out of [shelter] and my job was to house youth who came into the shelter.And, you know, back in the day when we would ask a landlord to rent to one of our youth or advocate, they would just be like, 'No I can't do it; they're on welfare'.Which legally you can't do.Well landlords have just got smart, and now they're just like-all landlords have to do is say is 'I found someone more suitable for my rental'.That's all they have to say.(Mike) In cases where youth (and their workers) are able to overcome the structural barriers to accessing rental housing in the private market (e.g., cost and discrimination), the housing that is available to them is often of low quality and/or within environments not conducive to their development and stability.As another HSW, Stephanie, put it to us: Youth that we're trying to keep out of shelter, [youth who do] not need the resources of a Housing First program, are still having to pay the same amount of money as these other folks.I have 16, 17-year-olds, and it's like, a room rental is all that you can afford.So, you're going to be living with like 18-plus-year-olds, and there's going to be substances around, there's gonna be alcohol, there's going to be people that you don't know.(Stephanie) Here we see that, for youth who may not need more intensive case management support, a scarcity of affordable housing limits the options available to them, which can lead to inappropriate housing solutions.
These same barriers impact the efficacy of shelter diversion programs, including efforts to quickly rehouse people who have had a tenancy compromised.As one shelter manager described, in response to a clarifying question about her use of the term rapid rehousing, there is nothing rapid about the work of securing housing for most homeless youth in Ontario right now: [Rapid rehousing] doesn't happen anymore.To be honest, we have always focused on making the experience of homelessness as short as possible.Do we get there?No.The shelter used to [aim to get people housed in] six weeks.Now we have stopped even using that term, because we know that we are not going to get anybody out of here in six weeks.So, ['rapid rehousing'] really means different things for who is coming in.For us, it means that we try as fast as possible to make that experience of homelessness short and get them out of this physical space as fast as possible.But really, it could be six months, it could be three months, it could be two days. . .we have become more and more unable to do rapid rehousing.(Noreen) In the face of these structural barriers (e.g., private landlord-market relations) and systemic inadequacies (e.g., government income supports), YITWs, HSWs, shelter workers, and managers are finding it increasingly difficult to rapidly rehouse or divert youth from the shelter system.In this context, workers are having to find creative ways to prevent experiences of homelessness on a case-by-case basis.

Working Creatively to Prevent Youth Homelessness
The flexibility of the YITW and HSW programs, in terms of what kinds of support and case management can be offered to youth, were described as important benefits of the program.A former HSW described how the flexibility of the program allowed them to meet clients where they were at, literally and figuratively: I would visit kids in their houses.Wherever they were staying I visit them.Wherever the heck they wanted.If I felt safe enough to do so.And Yeah.Whatever it is that they were dealing with at that time or at that period of their life is what we focused on.So, once I got them into a house.You know, what's our plan?What does this person need for you to be able to stay there; Is it 50 bucks a week to help with groceries?Is it that they need to be aware that they're working with a team?. . .there's a lot of mediation as well.(Darren) For some workers, this kind of flexible service delivery also means taking on a more assertive advocacy role for their clients.As Mike, a YITW, described: Our job is to hook them up and advocate for them with service providers in the community and try to get rid of some of that black tape, try to use my personal connections with people, to maybe bump their name up on the waitlist, things like that and open those doors.Things that they should be able to do on their own, but people or organizations sometimes just don't give them the time of day because of waitlists and things like that.So, we can be an advocate and really push our way through some of those things.(Mike) For Mike, this kind of advocacy is understood as an important benefit and service to be provided to the youth they work with.It lies in tension with the rationale for mandated data practices as described to us by one municipal data administrator: So, if we use this method and we use the By-Name List, then it doesn't matter if you've got 10 cheerleaders in your corner or no cheerleaders in your corner, if you are the person that statistically needs access to that service the most, then you're going to get considered first.So that's the whole purpose of the By-Name List, it's to ensure that there aren't side doors for people who maybe can engage more versus people who can't, for whatever reason, so to try and add a bit more equity into the system.And it's not just all about who's waving the loudest, saying "this is my client; this is my client!" (Emily) While Emily observed that data is the new advocacy in resource distribution contexts, other municipal administrators continue to see the value in more hands-on negotiated processes of case conferencing to determine program fit for clients.One municipal service manager (Valerie) described the value of case conferencing to us this way: I think there is a lot of value in us coming together.Again, because not all of that information is in HIFIS yet.And you learn so much about, you know, they've been through these three programs before that aren't actually documented anywhere.Or sometimes a case worker has reached out but there hasn't been an intake yet and that's not indicated.And it also does help us re-evaluate.Sometimes somebody will be referred to ICM but they're actually doing really well with an early intervention worker, and they don't need that level of resource.Or conversely, they seem like they have a lower acuity of need but then, staff working with them say, no there's actually all this other stuff going on; I think they would benefit from a higher level of resource.
While some managers, like the one quoted above, see the value of case conferencing within the Coordinated Access system as a whole, in the context of youth shelter diversion, case conferencing was described to us as a particularly important part of service provision.In multiple conversations, shelter workers described to us how diversion practices provide a means of working around statistically based housing allocation formulas.As one former shelter worker, Gertrude, described it: With our diversion work, that's where it was a bit of a-it was a more human process where we open it up to internally to [the Shelter] only; we didn't work with partners because it was just going to be far too complex for us to manage.And so that's where outreach workers would decide as a team like who are the three or four people being brought forward for that one bed and then my team continued with that process which was like interviews and engagement to ensure that folks were a good fit.[Gertrude] As identified here, and as described to us by a number of our interview participants, there is an unresolved tension between regimented data practices for housing allocation and the work of youth homelessness prevention which requires more flexible and adaptive service delivery models to effectively support youth.In the context of this ongoing tension, workers and systems managers are finding ways to creatively work around regimented systems to provide the best service possible for their clients.Sometimes this involves working against, rather than with, established data practices and service provision models because, as one shelter-based youth homelessness prevention worker stated bluntly: "HIFIS and the By-Name List do not support prevention work at all" (Jackie).

Discussion
Recent scholarship has noted a prevention turn in homelessness policy and research literature in the global north [31].Early homelessness responses focused on the provision of emergency services and late crisis intervention.While these responses are still present within the service provision ecosystem, they are being considered alongside approaches aimed at "homelessness prevention".For youth, whose development and long-term wellbeing are understood to be at risk of harm from homelessness, the case for prevention is all the more compelling.
In this context, scholars and homelessness advocates in Canada have proposed a homelessness prevention framework [20] and a roadmap for youth homelessness prevention [22].These tools are designed as guides to support policy makers and those working within the youth homeless-serving system with recommendations and program models to "(1) reduce the likelihood that a young person will experience homelessness, or (2) provide youth experiencing homelessness with the necessary supports to stabilize their housing, improve their wellbeing, connect with community, and avoid re-entry into homelessness" [22] (p.20).
Unfortunately, our research suggests that "the immediate provision of housing and supports to youth experiencing homelessness or housing precarity" [32] (p.9), a foundational requirement of youth homelessness prevention, remains difficult, if not impossible, to achieve in many communities.In the absence of government support for "universal" homelessness prevention strategies (e.g., a basic income guarantee) and given the scarcity of appropriate and affordable housing resources, the roadmap for youth homelessness prevention is challenging to use.In this context, the work of youth homelessness prevention in Ontario is being led by the frontline workers and managers (largely in the homelessness sector) who are frustrated and disheartened that despite their best efforts, young people are still spending long periods of time homeless before a suitable unit becomes available for them.
Though YITW and HSW workers have the capacity to provide personally tailored and flexible supports to an indicated population of youth (those involved in child welfare or self-declared as requiring protection), their work continues to be frustrated by structural barriers that cannot be addressed at the individual level.Shelter diversion holds promise for providing early intervention into experiences of homelessness by linking youth to natural or other supports prior to their first experience of homelessness.However, this work is often in tension with mandated data practices and prioritization processes which are oriented towards those experiencing chronic homelessness.Furthermore, in some cases, the ethics of frontline work (e.g., the protection of privacy) are in direct tension with the requirements of these government policy priorities (e.g., data collection).
From this vantage point, it becomes clear that while we may be witnessing a discursive turn to prevention, the day-to-day work of youth homelessness prevention in Ontario remains frustrated and forestalled by the wider social context in which these individual activities are deployed.This signals a clear need to attend to primordial prevention to address the "systems and policy level preconditions" necessary to end experiences of homelessness before they begin.It also signals the need to take seriously the empowerment of workers in the processes of service delivery to ensure that they can best support their client's access to safe and appropriate places to live.
We believe a more principled, rather than a technocratic or procedural, approach to conceptualizations of youth homelessness prevention is critical for ensuring that the ethos of homelessness prevention can be translated into transformative action.Practically, this looks like deep and sustained provincial investments in education, health, and social welfare-including income supports, housing, and public transportation.While service models and best practices may be sufficient to address some experiences of homelessness on a case-by-case basis, a principled prioritization of the right to housing (already enshrined within Canadian law) is needed to ensure the efficacy of youth homelessness prevention initiatives.