Housing First for Youth Who Experience Homelessness: A Systematic Review

Abstract

Housing First programs are widely used to house adults with histories of chronic homelessness. Recently, Housing First for Youth (HF4Y) emerged as a targeted response to youth homelessness and is presently cited as the ideal model; however, researchers have yet to synthesize the evidence on which this claim is made. Through a systematic review of literature, the present authors fill this gap by offering a thematic synthesis of available peer-reviewed evidence on the impacts of HF4Y on a variety of outcomes for youth who experience homelessness. Five databases were searched from inception to April 2023 and reference lists of relevant articles were hand searched for additional studies. This search finds seven studies that specifically measure outcomes associated with HF4Y enrollment. Findings indicate that HF4Y can improve housing security, but additional research is needed to develop a more robust evidence base. There is a need for longitudinal studies, randomized controlled trials, and rich qualitative studies of HF4Y.

Introduction

Homelessness is a persistent public health and human rights concern (Otto and Lynch 2004; Panter-Brick 2002). Youth who experience homelessness (YEH) are at increased risk of social, psychological, physical, educational, economic, and wellbeing challenges (Gaetz et al. 2016; Kull et al. 2019; Martin et al. 2006; Martijn and Sharpe 2006). They are more likely than housed youth to be victims of crime, exploitation, and abuse (Gaetz 2004; Madden et al. 2021; Middleton et al. 2018; Murphy 2016; Pilnik 2018). The longer youth remain unhoused, the more likely they are to experience these challenges (Rice, Stein, and Milburn 2008; Saade and Winkelman 2002), which indicates that there is an urgent need to assist YEH with housing.

Housing First programs are used widely across North America to provide expedited access to subsidized housing for adults who experience homelessness. These programs provide rapid access to housing that is separate from one's ability and readiness to access services for mental health, substance use, and education or employment, and evidence indicates that Housing First increases housing stability in adults (Woodhall-Melnik and Dunn 2016). While Housing First is described as an evidence-based approach for housing youth (Gaetz 2014), to date, there have been insufficient efforts to synthesize the evidence used to assert this claim (Youngbloom, Farnish, and Schoenfeld 2021). This gap is problematic as youth, as elaborated on elsewhere in this article, have unique needs that are not always best served by programs designed for adults (Brakenhoff et al. 2022). While the authors of this article are not the first to recognize this gap, as evidenced by growing attention to Housing First for Youth (HF4Y) data in the international arena and the publication of protocols for studying HF4Y in Canada (Gaetz et al. 2023), there remains a need to review the literature on HF4Y to better understand the efficacy of the approach.

The objective of this article is to assess the current state of knowledge on outcomes associated with participation in H4FY. Hence, this article answers the question: what broadly defined outcomes are found for youth participation in HF4Y? As this model is referred to as best practice, the authors seek to determine if the evidence is robust enough to make this claim when applied to the case of YEH. The secondary goal of this article is to indicate gaps in research on HF4Y that can be filled with future scholarship. Although this study aims to include any articles in English and French, regardless of geography, the studies uncovered are from the United States and Canada. This is not surprising given that Housing First emerged in North America. Hence, the background to this article focuses heavily on homelessness research and governance in the Canadian and American contexts. With this context in mind, this article now turns to a discussion of housing and homelessness in North America and then provides background on youth homelessness and Housing First. This background is followed by an overview of the methods used for the systematic literature review and its findings. This article concludes with a discussion of the findings and a brief conclusion.

Background

Homelessness in North America

In many ways, the focus on the United States and Canada within the literature makes for a more straightforward assessment of outcomes of HF4Y programs as there are significant convergences in the economic and political organization of the two countries. These convergences shape the landscape of housing affordability and the types of social programming available to those experiencing hardship. Specifically, both the United States and Canada are what Hall and Soskice (2001, 21) categorize as “liberal market economies” which, among many other characteristics, have higher rates of income inequality. Additionally, both Canada and the United States have federal systems of government (Simeon and Radin 2010) and relatively low social spending compared to other OECD countries (Béland et al. 2021). In the housing sector specifically, the United States and Canada have the lowest and second lowest public investment in social housing in the Western world, respectively (Bryant 2016). Both countries also have significantly financialized housing markets (August 2021; Khan 2024). Moreover, both countries rely on enumeration practices (e.g., point-in-time counts) to capture the depth of homelessness, including the enduing problem of youth homelessness. For both Canada and the United States, the official numbers likely underestimate the prevalence of youth homelessness as engaging YEH is often challenging (Dionne et al. 2023; National Conference of State Legislators n.d.). Available data demonstrates that homelessness among youth is present in both rural and urban spaces (National Conference of State Legislators n.d.), although Canadian researchers note that there is often migration to larger cities (called the “Migration Solution”) (Woolley 2016).

Youth Homelessness

YEH live in a variety of unstable, precarious, unsafe, and temporary housing situations such as couch surfing, “doubling up,” sleeping rough, staying in homeless shelters, or living in inadequate accommodations (Canadian Observatory on Homelessness 2019; Parker and Mayock 2019; Elkins, Farrell, and Fry 2023). While these patterns of housing are not necessarily different from adults, YEH need specialized supports at this crucial developmental stage of life, wherein individuals are typically supported as they learn skills to function independently (Tompsett, Fowler, and Toro 2009). It is a time when educational choices and entry into the workforce shape future outcomes (Schwan et al. 2018) and where brain development continues to occur (Konrad, Firk, and Uhlhaas 2013). During this stage of life, youth benefit immensely from relationships with supportive adults (Gasior, Forchuk, and Regan 2018). For example, Sieving et al. (2017) find that parental support and positive adult mentorship are associated with stronger developmental, social, academic, and health outcomes for YEH. However, these youth often lack these supportive nurturing relationships (Gasior, Forchuk, and Regan 2018). YEH not only experience additional precarity as a result of being unhoused, they also typically suffer from limited access to positive relationships with and guidance from adults. This lack of guidance, combined with a lack of access to economic and other social resources, contributes to worsening health, social, and economic outcomes as youth transition to adulthood.

The harms associated with youth homelessness are severe and multifaceted. YEH are at risk of victimization, sex and/or human trafficking (Wolfe et al. 2018), discrimination related to sexual orientation and/or gender identity (Abramovich 2012), self-harm, suicidal ideation, and suicide attempts (Stewart and Townley 2020). YEH often have prior experiences of trauma, such as placement in state systems of care, abuse, neglect, parental substance use and/or psychiatric disability, and a lack of parental acceptance or abuse due to sexual orientation and/or gender identity (Abramovich 2012; Dworsky, Napolitano, and Courtney 2013; Ferguson 2009; Wolfe et al. 2018). As a result of systemic, familial, and interpersonal trauma, youth may find it difficult to engage with adults in programs that provide supports and resources (Sapiro 2020) and they may distrust individuals and organizations that seek to provide help (Vitopoulos et al. 2017). Although youth often respond positively to trauma-informed approaches and peer support (Forrest et al. 2018), their ability to access or connect with services that use these approaches is limited by a lack of trust in systems, people, and services, which have consistently failed and traumatized them throughout their lives (Curry et al. 2021).

YEH often have complex traumatic histories and benefit from housing programs and policies that take trauma informed approaches and view safe and stable housing not only as a human right, but also as harm reduction (Kerman et al. 2021). However, quite often, housing programs focus on traditional trajectories associated with youth transitions to adult life, and hence require engagement with education and/or employment to access housing (Shelton 2015). This is problematic as it does not address the histories of trauma that accompany many YEH. It also fails to recognize the human right to housing and the need for stable housing to attend school and work (Sohn and Gaetz 2020). Further, youth may be viewed by service providers, government, and others as children who need supervision rather than mentorship, which often contrasts with their own views of self as they seek to establish independence (Thompson et al. 2006). Programs that are based on these views may not work for all youth and may be particularly ineffective for youth who experience trauma and strive for independence.

Housing First

Housing First as a philosophy and program model was developed by Tsemberis (Tsemberis and Eisenberg 2000), through the Pathways to Housing program in New York City. The philosophy and program model echo Maslow's (1943) hierarchy of needs, which states that to realize one's full potential and experience self-actualization, one must feel safe, secure, loved, connected, and confident. All of the feelings and stages within the hierarchy cannot be achieved without first meeting one's basic physiological needs for shelter, safety, clothing, sleep, water, and food. Housing First aims to remove all barriers to the right to housing (Collins and Stout 2020), through the rapid provision of subsidized housing to individuals who experience homelessness (Macnaughton et al. 2015). Individuals are provided with case management or Assertive Community Treatment (ACT) supports and access to a variety of other programs and resources that are optional and geared toward personal and vocational development, and recovery from addictions and treatment for mental illness (Gaetz, Scott, and Gulliver 2013; Collins et al. 2019). These supports vary by program and population served, but the basic premise remains that individuals should experience as few barriers to housing as possible.

The Canadian government formally adopted Housing First through the Homelessness Partnering Strategy in 2016 (Employment and Social Development Canada 2016). With this adoption, the federal government required communities with high levels of homelessness to allocate funding to Housing First initiatives. In 2019, the government removed this requirement (Government of Canada 2022), but the intensive investment during that period expanded Housing First programming nationwide. Housing First programs are also widespread in the United States, where the model originated. The approach has also gained popularity in European countries such as Finland, where it is known for effectively addressing long-term homelessness (Tainio and Fredriksson 2009), and France, which conducted a randomized control trial to assess the outcomes of Housing First participation (Aubry et al. 2021). Housing First operates alongside other programs to respond to long-term homelessness.

Housing First differs from other housing-led responses to homelessness, such as rapid rehousing and supportive housing programs, as it focuses on the provision of housing with support that is not time limited or dependent on meeting certain conditions (e.g., treatment for psychiatric diagnoses and/or substance use, employment, school enrollment, etc.; Woodhall-Melnik and Dunn 2016). Furthermore, there is a separation of housing from the provision of support and client choice in housing is emphasized (Gilmer et al. 2014). Housing First provides stable support and resources overtime, whereas rapid rehousing aims to provide highly flexible, time limited, immediate, housing assistance to stabilize individuals while they wait for sustainable long-term housing options (Byrne et al. 2023). Byrne et al. (2023, 616) draw on the work of Culhane, Metraux, and Byrne (2011) when they refer to rapid rehousing as a form of “‘secondary prevention’ from the field of public health [as it] seeks to reduce the impact and duration of homelessness by intervening and rehousing individuals as quickly as possible after the onset of an episode of homelessness.” When it comes to permanent supportive housing, some models follow the Housing First principle of separating housing services from other social service provisioning, including, importantly, dispersed models where tenants live in independent units in noncongregate settings. In other models, for example in high-support housing, other social service provisioning, such as daily programming, medical care, or medication supervision (Haley, 2017), may be built into the housing program and is then inseparable from the units or building.

Research demonstrates that Housing First programs that maintain high fidelity to the intended program model have higher housing stability rates (Davidson et al. 2014; Gilmer et al. 2014). However, programs are often constrained in their ability to meet these mandates. Pauly et al. (2013) note that without access to an adequate number of safe, affordable, and appropriate housing units, the model cannot comprehensively meet the needs of all those who might benefit from a harm reduction approach to housing. Most pilot programs and those that are dependent on government funding cycles lack sustainable funding for programming and rents, making access to housing temporary (Stadler and Collins 2021). Further, despite years of study, some of the long-term benefits of Housing First, and short-term, nonhousing outcomes, are not yet established (Baxter et al. 2019). The short-term housing benefits of Housing First are well-established for certain populations (e.g., adults who experience chronic homelessness), but short-term outcomes for other populations (e.g., youth, survivors of intimate partner violence, etc.) have yet to be systematically examined (Verdouw and Habibis 2018). This indicates a need for additional study of the outcomes associated with Housing First programs across a variety of populations and contexts.

Housing First for Youth

According to Gaetz, Walter, and Story (2021, 7) “HF4Y is an adaptation of the well-established Housing First approach to addressing homelessness.” HF4Y recognizes that the causes and consequences of youth homelessness are different from those of adult homelessness and, therefore, responses must be tailored to meet the specific needs and concerns of youth. Gaetz, Walter, and Story (2021) and Gaetz et al. (2023) argue that although HF4Y is adapted from the adult model, there are distinctions in program principles, goals, outcomes, and case management. They describe five guiding principles of HF4Y which state that providers must:

Recognize the right to housing with no preconditions;

Emphasize youth empowerment and choice;

Focus on youth wellness and development;

Provide individualized supports that are not time limited; and

Promote social inclusion and community integration.

To stay in HF4Y programming, youth must meet the two following conditions: (1) weekly meetings or contact with case management and (2) if youth have income, 30% must be contributed to their rent (Gaetz, Walter, and Story 2021). Gaetz, Walter, and Story (2021) specifically note that adult programs that serve youth over the age of 18 are not HF4Y, as programs must be intentionally designed to ensure that resources are tailored to the needs and developmental stage of youth. For example, they recommend that programs use trauma-informed care approaches and incorporate principles of positive development, such as goal setting, self-esteem building, and the provision of vocational support. Like adult programming, the success of HF4Y depends on the ability of programs to adhere to fidelity measures that dictate the key tenets and requirements of Housing First programs (Gaetz, Walter, and Story 2021).

Evaluation reports of HF4Y programs are predominantly found in gray literature and demonstrate varying degrees of success (McParland et al. 2019). However, the peer-reviewed literature base has yet to be reviewed to determine whether HF4Y demonstrates successful outcomes. The present authors identify five reviews that focus specifically on the outcomes of youth enrollment in various housing interventions (see Altena, Brilleslijper-Kater, and Wolf 2010; Morton et al. 2020; Slesnick et al. 2009). The reviews conducted by Altena et al., Barker et al., and Slesnick et al. are considered outdated by researchers who discuss best practices for updating systematic reviews (see Garner et al. 2016). Further, the use of HF4Y was not common prior to the publication of these reviews. Morton et al. (2020) synthesize evidence published up to March 2019 and include both housing and nonhousing outcomes, such as mental health and wellbeing. They conclude that there is a paucity of evidence that evaluates the effectiveness of program models to address youth homelessness in general. However, their work is not specifically focused on HF4Y. The final review, conducted by Slesnick et al. (2023) explores literature on youth outcomes from rapid rehousing, permanent supported housing, and HF4Y. However, this review—published after the submission of the present review—does not specifically isolate and discuss the implications of HF4Y and instead examines all three housing-led interventions.

Methods

This manuscript presents a systematic review of peer-reviewed evidence on outcomes associated with HF4Y programs. The authors followed the Joanna Briggs’ Institute (JBI) model for systematic reviews (Institute TJB 2014). JBI reviews are often used by health researchers, with the goal of working with available evidence to establish evidence-based practices for healthcare (Institute TJB 2014). The JBI model is particularly useful for producing evidence that practitioners can use when determining if an intervention is appropriate, feasible, and effective (Pearson et al. 2005). Although Housing First is not traditionally viewed as a healthcare intervention, housing is a social determinant of health (Hernandez and Suglia 2016) and, as noted above, a lack of access to suitable and affordable housing is well-understood as a contributor to diminished wellbeing in youth.

Search Strategy

A librarian (AG) developed the search strategy in consultation with the authors. Searches in PubMed, Google Scholar, PROSPERO, and the Open Science Framework (OSF) confirmed that no reviews on HF4Y had been previously conducted or reported as in progress. To begin developing the search strategy, AG identified two searchable concepts from the review question, “youth” and “Housing First,” and gathered potential search terms for each concept from terms provided by the authors and terms located in relevant database records. This was done through an exploratory search of MEDLINE (Ovid) and CINAHL (EBSCOhost) and involved analyzing the words contained in the titles, abstracts, and subject headings of database records about youth and/or Housing First and adding new terms to the list of potential search terms for each concept. Terms for youth included adolescent, adolescence, teenager, teen, youth, juvenile, emerging adult, young adult, homeless youth, vulnerably housed youth, and high schooler. To ensure that applicable articles that examined outcomes of multiple youth housing programs which included Housing First were not overlooked, terms for housing types included Housing First, permanent supportive housing, intermediate housing, pathways to housing, supportive housing, stable housing, barrier free housing, transitional housing, temporary housing, housing intervention, affordable housing, and subsidized housing. Terms for specific Housing First initiatives such as Housing First, Shelter plus care program, At home/chez soi, Four wise men, and Un chez-soi d’abord. To identify potential articles that compared Housing First to other programs, Foyer was included as search term, as it is a prominent non-Housing First program for YEH. Subject headings were also retrieved for each concept. These were tailored to each database (e.g., homeless youth and adolescent in MEDLINE, homeless persons, adolescence and young adult in CINAHL). Additional subject headings were located when the search was adapted to each database, such as youth and teenagers in Academic Search Premier.

The librarian (AG) then created a draft search strategy using the identified text words and subject headings. Each term was searched in MEDLINE (Ovid) to determine whether the search results reflected the scope of research available relevant to the review question. This process involved searching each individual term within and against the concepts to ensure that only terms which returned unique and relevant results were included in the final search strategy (see Table 1 for an abbreviated strategy and Supplemental Appendix A for additional details). AG then reviewed a sample of the search results to determine whether records that were relevant to the review question were located as each new search term was added. If a term produced relevant results, it was added to the final search strategy. This process continued until all identified text words and subject headings were searched. Hence, not every identified term was used in the final search strategy. For example, the term juvenile did not return unique, relevant results in any database and the term emerging adult was not useful in every database. Once all the identified terms were searched, the draft search strategy was reviewed by a second librarian (RW) using the Peer Review of Electronic Search Strategies (PRESS) guidelines (McGowan et al. 2016). Recommended adjustments from the PRESS were made to the strategy, such as adding subsidized housing as a text word.

Table 1.

Simplified List of Search Terms and Databases.

Databases	Participants	Housing Interventions
Ovid MEDLINE; EMBASE (Elsevier); APA PsycINFO (EBSCOhost); CINAHL with Full-text (EBSCOhost); Academic Search Premier (EBSCOhost)	Adolescent; adolescent* homeless youth; teen*; young adult; youth adult*; homeless persons; young adult; young adult*; emerge* adult*	Supportive housing; stable housing; permanent housing; transition* housing; temporary housing; subsidized housing; affordable housing; public housing; Housing First; Pathways to Housing; At Home/Chez Soi; Foyer; housing program;

The final search strategy was created by combining subject headings that were unique to MEDLINE with text words which were truncated as needed (e.g., shortening adolescent to adolescen* to capture adolescence, adolescent, and adolescents). Database search fields for title/abstract/keyword were also used for searching text words. For example, MEDLINE text words were searched using title/abstract/keyword heading/keyword heading word fields represented by the search fields ab, kf, kw, ti whereas in APA PsycINFO text words were searched in the title/abstract fields TI OR AB (see Supplemental Appendix A). This approach was taken to limit the databases to only retrieve records where the text words appeared in the selected search fields. This ensured that results were more relevant to the review question. All terms were combined using Boolean operators and syntax that was appropriate to MEDLINE. The search strategy was considered finalized in MEDLINE, as all previously identified terms had been searched, with no new or meaningful additions to the search results and no new terms identified.

The search strategy formed in MEDLINE was adapted to the specific syntax and subject headings that were appropriate to the following five databases: MEDLINE (Ovid), Embase (Elsevier), CINAHL with Full-Text (EBSCOhost), APA PsycINFO (EBSCOhost), and Academic Search Premier (EBSCOhost). This process involved using the text words and subject headings from the MEDLINE search strategy as the basis for determining whether the same text words returned unique, relevant results and reflected the scope of research captured in each database, or whether other terms needed to be added or removed from the list. Subject headings unique to each database were also located and included, as appropriate, in the search strategy for each database. Search strategies for every database were combined using Boolean operators and syntax appropriate to the database and were finalized when all identified terms had been searched with no new or meaningful additions to the search results. No limits were applied to the search results in any database, and databases were searched from their inception to April 18, 2023 (see Supplemental Appendix A for the full search strategies for each database). The authors acknowledged that several well-done gray literature reports have been written on HF4Y. These reports have also been reviewed and included in the literature review and discussion sections of this manuscript. However, as these reports have not undergone a blind peer-review process, they are not included in the results of this systematic review.

Inclusion and Exclusion Criteria

To be included in the review, papers had to focus on any outcome (broadly defined) for YEH enrolled in programs defined in the articles as HF4Y, permanent supportive or supported housing for youth that follows Housing First principles. As noted above, permanent supportive and supported housing does not require that housing be separate from supports; however, in some cases, as in those that follow Housing First principles, it is. Homelessness was defined as unhoused or precariously housed and unaccompanied by parents, caregivers, or guardians. An open-ended operationalization of youth was used, and search results included programs with age ranges between 18 and 36 years. This was intentionally defined broadly, which allowed for typical variations in the definition of youth used by different jurisdictions and programs.

All included papers measured or used data to describe at least one outcome of a HF4Y program. Qualitative, quantitative and mixed methods studies were all appropriate for inclusion, so long as they focused on outcomes associated with participation in HF4Y. Due to the paucity of literature on this topic, all outcomes measured were included. Further, not all papers provided clearly defined labels for the housing interventions studied, which required the authors to make some decisions about what constituted Housing First. As the main premise of Housing First is to provide low-barrier access to housing with few or no preconditions (Woodhall-Melnik and Dunn 2016), the studies included had to investigate programs that had no preconditions (e.g., enrollment in school, participation in employment, abstinence from substances, etc.) for youth to receive housing. Studies that investigated rapid rehousing or provided permanent supportive housing with requirements for enrollment in education, employment, training, abstinence or mental health treatment were excluded, as these types of interventions do not follow the basic premise of Housing First. These studies were marked as “wrong type of intervention.” This resulted in the inclusion of four papers (Brothers et al. 2020; Dodd, Ruffins, and Ruffins 2018; Gilmer 2016; Youngbloom, Farnish, and Schoenfeld 2021) that studied programs that were described as permanent supportive housing with Housing First attributes, as they met the criteria of providing housing without preconditions and the separation of services from housing.

The geographic and temporal context of the included literature was unrestricted. Study protocols, literature reviews, commentaries, and methods papers were excluded from this review. Papers published in languages other than English or French were excluded as they could not be reliably reviewed by the authors. Studies that did not evaluate or assess the impact of HF4Y on youth outcomes were excluded. Further, papers that discussed outcomes related to youth participation in adult Housing First programs were removed, as Gaetz, Walter, and Story (2021) note that these are not representative of HF4Y, as they are not specifically designed to meet the needs of youth. This resulted in the removal of two articles from the At Home/Chez Soi study (Kozloff et al. 2016a, 2016b). The authors chose to keep the article by Collins et al. (2019) in the analysis, although the program was designed for both youth and families with children. This decision was made at the full-text review stage, with the rationale that the program considered the needs of youth in its design and reported separate youth outcomes.

Data Extraction

Following the search, all identified citations were collated and uploaded into Zotero reference management software and then into Covidence systematic review management software. Duplicate references were removed in Covidence. Two independent reviewers from the research team screened all the titles and abstracts for assessment against the inclusion and exclusion criteria. Full-text articles were retrieved from all potentially relevant sources and three members of the research team reviewed each of the full-text documents to ensure that they met the study criteria. Any disagreements that arose between the reviewers at each stage of the selection process were resolved through discussion.

The inclusion and exclusion process at the title and abstract screening, full-text review, and extraction phases were presented in a Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRIMSA) diagram (see Figure 1). PRIMSA is commonly used in systematic reviews to standardize procedures and increase the transparency of reporting of reviews (Moher et al. 2009).

Figure 1.

PRIMSA diagram.

Data were extracted from the papers by three authors who performed the full-text reviews. Data extraction was tailored to answer the review question: what broadly defined outcomes are found for youth participation in HF4Y? To answer the review question, the authors extracted data on (1) the outcomes measured and study findings; (2) gaps in services offered to youth; (3) research methods; and (4) gaps in research and knowledge. The data extraction also included specific details about the study participants and programs (e.g., age range of youth served, types of services offered, etc.). The data extraction chart was then reviewed by the lead author for general patterns and themes and then condensed into a smaller chart for inclusion in this manuscript (see Supplemental Appendix B). To present the findings, we have focused on the main outcomes of the studies, discuss the strengths and challenges of HF4Y, alongside recommendations for improvement, and discuss the need for further evidence that was presented in the reviewed studies.

Limitations

The present review has several limitations. Some of the articles did not provide comprehensive descriptions of programs or referred to themselves as supported or supportive housing programs for youth with Housing First attributes. The authors of the present study included these articles based on their review of the program contexts provided in the articles. However, these articles did not assert that they were HF4Y. The articles which did not clearly define themselves as HF4Y but appeared to have very similar attributes are indicated in Supplemental Appendix B. Despite this limitation, the present authors argued that removing these studies would not have changed the finding that there was a paucity of research on HF4Y and that more research that focuses on appropriate services and supports for youth is needed.

Findings

The initial search found 3,707 articles, of which 1209 were duplicates, resulting in the inclusion of 2,498 studies in the title and abstract review. Eighty-five were included in the full-text review, seven of which were included in the present review. Six articles that were deemed of potential interest were not accessible through the authors’ library systems. Four of these were published before the creation of the HF4Y model (published in or before 1995) and the remaining two were unavailable for document delivery. The primary reason for exclusion was that articles did not address outcomes of youth participation in Housing First (N = 83). In other words, they focused on youth housing solutions that were not HF4Y. Additional exclusions were made of articles that were focused on youth participation in Housing First but did not specifically study youth models (N = 2). The authors conducted a hand search of the reference lists of the included studies to locate potential missed sources. They also examined the articles that cited the included articles. No other relevant papers were found. The results of the data extracted from the seven papers are presented thematically below.

As expected, there was variation in the ages used to define youth, which ranged from 18 to 36 years old. However, all but one of the articles focused on individuals aged 18 to 24 (N = 6). The one article with a broader age range included youth aged 18 to 36 (Guevara et al. 2019). All seven studies were about HF4Y programs in the United States (N = 6) and Canada (N = 1). Themes uncovered through the full chart review were housing security, costs and return on investment, health outcomes, food security/access to food security programs, program fidelity, youth engagement with programming, program strengths, gaps and recommendations, and future research directions. Overlap was observed in some of the thematic categories, and hence, some categories were further collapsed for presentation in this manuscript. While the literature is limited, there is a need to conduct a systematic review given that Housing First models are being developed for use with youth.

The included studies investigated a diverse range of housing programs. Some authors noted that they drew on large databases wherein youth resided in multiple different housing programs that were HF4Y, supportive or supported housing programs that followed Housing First principles, whereas others drew small samples from isolated housing programs. Methods varied greatly and included analyses of linked administrative data, quantitative analyses of participants’ program records, surveys, participant observation, in-depth interviews with program staff and other professionals, analyses of informal feedback from program participants and landlords, and return on investment analysis. Only one study (Brothers et al. 2020) included youth voice on HF4Y and participants’ perspectives were missing from the evidence base.

The diversity of methods used across the articles resulted in findings that could not be compared using meta-analysis. Further, the youth included in the studies had different characteristics, which made comparisons difficult. For example, Dodd, Ruffins, and Ruffins (2018) explored HF4Y with young adults diagnosed with HIV, whereas Guevara et al. (2019) studied outcomes in youth who experienced psychosis. Their findings reported important gaps in knowledge on HF4Y with specific morbidities; however, they were not generalizable to broader populations of YEH experiencing homelessness.

Housing Security

Four of the seven articles directly assessed housing security. In the studies, housing security was operationalized as the percentage of days spent in stable housing (Kelleher et al. 2021), the proportion of clients who maintained the initial housing unit received through the HF4Y program (Collins et al. 2019; Guevara et al. 2019), and the number of youth who left housing (Guevara et al. 2019; Youngbloom, Farnish, and Schoenfeld 2021). Kelleher et al. (2021) demonstrated high levels of housing security. They calculated the mean percentage of days housed at baseline and at follow up and found that rates improved drastically from 8.06% to 92.88%. The other three studies demonstrated complexity surrounding housing security. Collins et al. (2019) found that just over one-fifth of youth had returned to shelter after accessing the program and half were rehoused after losing their initial housing placements. Housing loss was related to poor quality or unsafe housing units, issues with landlords, and personal crises (Collins et al. 2019). Guevara et al. (2019) found that approximately 54% of youth with psychosis maintained their housing with the program. Youth who did not maintain housing either returned to their parents’ home, moved to supervised housing, accessed government subsidized housing, went to a shelter, or disengaged from the program. However, the sample size was small (n = 24), therefore, larger patterns in housing retention were difficult to estimate, and the authors note that their findings are not generalizable to all YEH. None of the studies factored in jurisdictional variations in tenant protections, rent control regimes, or other factors related to constrained housing markets that often contribute to evictions in low-income populations.

One of the four studies that assessed housing security examined patterns associated with housing loss. In this study, Youngbloom, Farnish, and Schoenfeld (2021) found that 40% of youth in HF4Y experienced housing loss at some point during program involvement. They systematically tracked evictions and noted that the majority (60%) came from issues with guest management (e.g., housing guests who were not on the lease). Other reasons for housing loss included incarceration, domestic violence, drug use, late payment of rent or fees, and psychiatric disability/dx. They used logistic regression to identify specific groups who were at greater risk of housing loss than others and found that youth with previous foster care involvement were 13.74 times more likely to lose housing than other youth in Housing First. LGBTQ + youth were 5.81 and youth with depression were 4.14 times more likely to lose housing than other youth enrolled in Housing First.

Recommendations to improve housing security were included in some of the articles. Youngbloom, Farnish, and Schoenfeld (2021) noted that youth at greater risk of housing loss may have benefited from resources that encouraged the development of strong, positive, supportive networks, and that facilitated connections to LGBTQ + community resources. They recommended that programs ensure access to safe spaces, and that strategies for youth connections to agencies and supports be investigated (Youngbloom, Farnish, and Schoenfeld 2021). Further, Collins et al. (2019) noted that Intensive Case Management was important for housing outcomes, and that guidance, focused on communication and parenting skills coaching and systems navigation, was most impactful for youth.

Youth Engagement With Programming

Youth engagement of with optional programming was considered in three of the studies. Guevara et al. (2019) found that 32% of youth were employed or in school when they entered HF4Y. Throughout their engagement with the program, 71% of youth maintained or resumed education or employment. At the time of enrollment, 35.5% had disengaged from services offered through the program; however, all youth re-engaged with services once housed. Collins et al. (2019) also measured educational and employment services accessed. They found that only the most stable clients were able to explore educational and employment options and that others who were less stable did not investigate options for school or work.

Kelleher et al. (2021) tracked program participation by type. They offered evidence-based programs that were specifically designed for youth and found that participation in optional programming was generally high. The average number of sessions attended per youth was 13.57 for self-advocacy services, 1.33 for motivational interviewing, and 0.76 for HIV prevention services. They do note that services were offered virtually due to COVID-19 restrictions, which may have impacted youth willingness to access services.

Collins et al. (2019) stressed the need to carefully tailor HF4Y programs to the developmental stages and needs of youth. They emphasized the importance of employing well-trained, compassionate, consistent, and flexible case management staff that are supported in their work.

Health Outcomes

Healthcare access and health behaviors were studied by both Dodd, Ruffins, and Ruffins (2018) and Kelleher et al. (2021). Dodd, Ruffins, and Ruffins (2018) found that all HF4Y participants accessed healthcare providers for HIV monitoring and treatment. Youth benefited from lower viral loads and fewer hospitalizations. Kelleher et al. (2021) explored substance use and mental health among HF4Y residents and found a significant decrease in cognitive distortions and reductions in marijuana use. Alcohol use remained unchanged. However, the consequences (e.g., legal, social, housing, etc.) related to substance use decreased significantly.

Food Security and Social Assistance

Two of the studies explored food security. Brothers et al.'s (2020) study focused specifically on food security among youth in a supportive housing program with Housing First attributes. They found that food insecurity persisted after housing was secured. This was particularly problematic near the end of the month as money ran out. They noted that the housing was in a food desert with limited access to fresh and affordable food. Further, youth in the program were housed in congregate housing and most used a shared kitchen, which facilitated food security for some individuals who shared food resources, but also contributed to food insecurity for others who felt pressured to share their limited resources. Some youth fared better with the communal kitchen than others. Fear of food or kitchen item theft was a barrier to kitchen use and food security for some, and lack of tools and cooking knowledge was a barrier for others. Brothers et al. (2020) found that youth coped with hunger with maladaptive strategies, such as using drugs, ignoring hunger, getting food from garbage bins or stealing. They recommended instating food delivery programs. They also recommend that HF4Y programs that are in food deserts focus more on the food provision. They also recommended life skills training with a focus on cooking and budgeting. They concluded that some youth may do better in individual apartments with their own kitchens, whereas others may do better in communal settings.

Collins et al. (2019) find that access to the Supplemental Nutritional Assistance Program (SNAP) and social insurance for income support increased for youth participants. Brothers et al. (2020) also found increased rates of social assistance use, but they found that youth were still hesitant to use food banks and other forms of food security programs, as they felt that there was stigma associated with food programs. To counter this, Brothers et al. (2020) recommended that youth be given access to food stamps with higher maximums.

Child Welfare Involvement

Collins et al.'s (2019) study was the only one to assess child welfare involvement as an outcome of participation in HF4Y. The program studied provides services to both families and youth. They noted decreased involvement with child welfare; however, the most significant decreases were experienced by families in the program, not youth.

Program Satisfaction

Program satisfaction was a strength presented by Guevara et al. (2019), who found high levels of satisfaction among HF4Y participants and the landlords who housed them. Youth were happy with increased housing security, autonomy, and life skills. Landlords, on the other hand, enjoyed consistent rent payments, and many had gained more positive views of individuals with psychiatric concerns. However, satisfaction was assessed through informal conversations with youth and landlords and details on how the authors demonstrated methodological rigor in collecting informal feedback were not provided.

Costs of Participation in HF4Y

Two of the four articles measured costs of HF4Y. Dodd, Ruffins, and Ruffins (2018) performed a Return on Investment (ROI) analysis and found that the ROI for HF4Y who lived with HIV was 1.32, with an estimated savings of $6,750 per year compared to traditional responses to the typical services provided to youth through the shelter system. They noted that their ROI increased to 1.43 when hospitalization rates were included in their calculations. They also noted that viral loads and hospitalizations were significantly reduced for those enrolled in HF4Y.

Gilmer (2016) investigated service costs for youth with significant psychiatric diagnoses who accessed permanent supportive housing programs that followed Housing First mandates. They found that the cost of providing mental health services for HF4Y was, on average, $13,337 USD higher for youth in the housing programs when compared to those who were not in the programs but who received public mental health services. These results indicated a need to assess the effects of HF4Y across different populations in a variety of contexts.

Importance of Fidelity to Housing First Principles

Only one article measured program fidelity in HF4Y programs. Gilmer (2016) found that fidelity to the HF4Y principles varied across programs. Those with higher fidelity had significantly fewer inpatient and more outpatient services accessed for mental health than those with lower levels of fidelity. Based on these findings, Gilmer (2016) concluded that not all programs evaluated were designed and implemented to meet the needs of youth and that fidelity to the intended HF4Y model was of upmost importance. Gilmer (2016) concluded that in order to maximize benefits, service providers need to design and implement permanent supportive housing programs with high fidelity to HF4Y principles.

Need for Additional Program Resources

Youngbloom, Farnish, and Schoenfeld (2021) argued for increased support and resources, as well as peer connections and social support. As they investigated the risk of housing loss, they tailored their recommendations to the realities of youth who were most likely to lose housing. They noted the need for more social support for all three higher risk groups of youth, with a specific focus on connections to LGBTQ + community resources and agencies for queer youth. They also recommended creating safe and inclusive spaces within HF4Y and suggest that future research explore strategies to counter help avoidance and social disconnection.

Guevara et al. (2019) noted a need for longer subsidies and sustainable program funding to ensure participants’ access to housing for as long as needed. They also described the need for peer support models in the program. Dodd, Ruffins, and Ruffins's (2018) recommendations aligned with Collins et al. (2019) in that they advocated for more resources and funding for permanent supportive housing programs that follow HF4Y principles. They stressed that it was important for social workers and healthcare providers to advocate for supportive housing programs.

Future Research Needs

The academic evidence base on HF4Y is small, which indicates a need for more research. Most researchers provided suggestions for future studies. Collins et al. (2019) discussed the need to collect more systematic data on client moves, evictions, and employment. They also noted a need for more studies that include the voices and perspectives of youth engaged with Housing First. This was particularly insightful, as there was a paucity of qualitative descriptions of lived experiences in the evidence base. More research was recommended to establish best practices for support services that met the unique needs of youth (Gilmer 2016).

Many of the researchers argued that there was a need for a randomized controlled trial or a pragmatic cluster randomized trial of HF4Y (Collins et al. 2019; Guevara et al. 2019; Kelleher et al. 2021; Youngbloom, Farnish, and Schoenfeld 2021). Ideally speaking, these studies would make use of control groups to establish generalizability and attribute findings directly to HF4Y participation (Collins et al. 2019). Kelleher et al. (2021) noted that this would allow researchers to draw more robust conclusions about the effectiveness of HF4Y. In addition to trial studies, Youngbloom, Farnish, and Schoenfeld (2021) called for longitudinal studies that examine the housing trajectories of youth over time.

Discussion

The present review highlights a paucity of evidence on HF4Y. This is surprising as there is a plethora of these programs in North America. The seven studies that meet the inclusion criteria explore housing security, economic costs, program strengths and gaps, fidelity to the HF4Y model, food security, social assistance rates, child welfare involvement, health outcomes, engagement with optional programming, and research gaps. A variety of methods and sample sizes are used and some of the program descriptions of the interventions were richer than others. The included studies demonstrate mixed outcomes on the effectiveness of HF4Y on a variety of different indicators including housing security, food security, costs, education and employment. Despite not limiting the search to North America, all of the included studies are on programs based in North America. This is likely due to the popularity of Housing First approaches in Canada and the United States. The authors of this review conclude that H4FY can lead to the development of some promising interventions for YEH; however, more research is needed on fidelity measures and on the development of youth-specific programming.

The findings of this review indicate that YEH have complex housing trajectories that are not necessarily linear or patterned. This is evidenced by the number of youth who leave their initial housing placements (Collins et al. 2019; Guevara et al. 2019), which suggests a need to be flexible and rehouse youth when housing is lost. The ability to be supported when making mistakes and learning is highlighted in the HF4Y philosophy (Gaetz, Walter, and Story 2021) and multiple chances for housing security are also a component of Housing First for adults (Johnsen and Teixeira 2012); however, this review only uncovered one study that measured program fidelity, which indicates that adherence to program principles is not well-understood. Further, the studies indicate that measures of retention rates for initial apartments may not be the most useful outcome to track the success of HF4Y programs. Many of the papers argue for the need for longer-term investigations of HF4Y. Future studies should track housing movement over time, paying attention to trends in the proportion of days spent housed, as measured by Kelleher et al. (2021), to establish more comprehensive understandings of housing security.

This review uncovers the need for additional, tailored supports for youth. For example, food insecurity persists past the receipt of housing (Brothers et al. 2020), which highlights the importance of housing location, as food security is particularly difficult to achieve in food deserts (Dahle 2021). Research consistently shows a correlation between access to fresh food sources and health (Algert, Agrawal, and Lewis 2006; Michimi and Wimberly 2010; Wang and Qiu 2016). Further, as other authors also find (Bowen and Irish 2018; Brito-Silva et al. 2022; El Zein et al. 2018, Johnson, Drew, and Auerswald 2020), low-income youth are reluctant to access food banks and other free or cheap food sources due to perceived stigma. This suggests the importance of being selective about neighborhood characteristics for youth housing. When choosing housing for HF4Y participants, attention should be paid to the availability of transportation, high quality and affordable food sources, and other resources to improve overall health and wellbeing.

The findings indicate that youth in Housing First access physical (Dodd, Ruffins, and Ruffins 2018) and mental healthcare (Gilmer 2016). Although Gilmer (2016) observes increases in service costs for mental healthcare, investment in more expensive healthcare resources in the short-term may help lower future expenditures. Further, mental health treatment may significantly lower healthcare costs as youth age (Colizzi, Lasalvia, Ruggeri 2020). Hence, the cross-sectional data that display differences in costs may not provide a good indication of the true economic contributions of HF4Y. Costs should be tracked for longer periods of time using administrative data linking to assess healthcare use and the presence of chronic disease.

This review highlights the importance of social connections for YEH (Barman-Adhikari et al. 2016). Like other studies (Dang and Miller 2013; de la Haye et al. 2012; Oliver and Cheff 2014), Youngbloom, Farnish, and Schoenfeld (2021) find that youth benefit from supports that focus on building healthy social networks and connections. Like other scholars, they find that youth with histories of foster care involvement (Huang et al. 2018), LGBTQ + youth (McDonald 2018; Rosario, Schrimshaw, and Hunter 2012), and youth with depression (Tyler, Schmitz, and Ray 2018) need more positive social support. Further, youth who become stably housed may disengage from former street or drug-involved networks (Kelleher et al. 2021) and become socially isolated (Kidd et al. 2019). Studies find that levels of support in youth are typically low to begin with, regardless of housing status (de la Haye et al. 2012; Stewart et al. 2010) and more emphasis should be placed on building positive peer connections and genuine mentorship among youth engaged with HF4Y.

Thoughtful, evidence-based approaches to youth-specific programming, demonstrated in Kelleher et al.'s (2021) study, appear to produce good youth engagement with services. In their study, youth are highly engaged with services and this could be related to the intentionality in the program's selection of supports that are specifically designed for youth. In outlining the principles of HF4Y, Gaetz, Walter, and Story (2021) and Gaetz, Walter, and Borato (2021) note that programs must be designed to meet the unique developmental needs of youth. Youth with histories of homelessness experience systems, supports and service engagement differently than their adult counterparts (Thompson et al. 2006). This review indicates that programming for youth should be created or chosen thoughtfully, with consideration paid to the established evidence-base on youth programming.

There are several gaps in the literature that should be addressed. First, it is unclear as to which type of housing best suits the needs of youth. Brothers et al. (2020) argue that congregate living works better for some youth, whereas others may fare better in individual apartments. Somers et al. (2017) find that adults in Housing First maintain comparable levels of housing stability in scattered site and congregate housing. However, they find that individuals in congregate housing experience more improvement in health and psychosocial outcomes. As youth have different needs than adults, more attention should be paid to which type of housing works for different youth with diverse experiences. Accessible evidence on this would provide programs with the ability to preemptively provide youth with careful guidance on housing choices and may increase housing stability in initial housing placements.

The studies in the present review explore outcomes for youth aged 18 years and older. This is problematic as other studies identify the average age of first entry into homelessness as younger than 18 (Martijn and Sharpe 2006; Rosario, Schrimshaw, and Hunter 2012). The developmental stages and needs of youth vary greatly by age. Research should aim to understand the outcomes associated with HF4Y in younger youth, as outcomes and experiences may vary greatly with age. Further, this review finds a dearth of youth voice in the evidence base for HF4Y. Incorporating lived experience when studying housing is immensely important to understanding the rationale for housing trends, trajectories, and solutions (Grover and Rappaport 2021; Nelson et al. 2014; Padwa et al. 2023). Qualitative research on Housing First for adults underscores the importance of understanding the actual experiences of individuals, to garner comprehension of the conditions and concerns that lead to stability and instability within the context of housing programs (Padgett 2007; Woodhall-Melnik et al. 2017). This context is important for program development, augmentation, and improvement. Future studies should focus on understanding youth experiences with HF4Y to better determine if these programs meet youth needs from their perspectives.

Only one study assesses fidelity in HF4Y programs and finds it important to housing stability (Gilmer 2016). Experts conclude that maintaining fidelity to the principles of Housing First, while adapting programming to the unique needs of specific populations, is fundamental to success (Aubry et al. 2015; Gilmer et al. 2014; Nelson et al. 2014). In presenting the principles of HF4Y, Gaetz, Walter, and Story (2021) and Gaetz, Walter, and Borato (2021) also stress the importance of program fidelity. Housing providers that start HF4Y programs should become familiar with what constitutes fidelity to overarching HF4Y principles. However, systematic tools for measuring HF4Y fidelity are needed and should be developed. Mentorship and training from well-established HF4Y providers on fidelity to the principles of HF4Y could be offered to service providers who establish new programs.

Finally, the majority of studies conclude that more systematic, comparative, and longitudinal research is needed on HF4Y. The authors of this review agree with this conclusion and note that at present, the evidence base for HF4Y is not yet robust enough to draw conclusions that this is the gold standard model for providing services to youth with histories of homelessness. However, this review indicates that HF4Y demonstrates promise for achieving housing security and supporting youth. It is important to note that barriers, such as low-landlord participation and lengthy wait times for subsidized housing may persist (Brakenhoff et al. 2022; Canham, Wister, and O’Dea 2019; Schwan et al. 2018), and negatively impact the capacity of programs to respond to housing needs. However, none of the articles provided an in-depth analysis of unit availability, tenants’ rights, rent control regimes, or the general landscape of social welfare provision in the jurisdictions that house the HF4Y programs. This is important to consider as these factors have a large impact on the ability of housing programs to meet the needs of their participants (Stadler and Collins 2021). However, additional research on HF4Y may provide a stronger evidence base which can be mobilized to meet the needs of youth.

The findings of the present review reinforce the findings from previous reviews conducted on housing interventions for YEH. For example, past reviews argue that there is a need for more methodologically rigorous research (e.g., experimental designs, longitudinal research, etc.) on housing interventions for youth (Altena, Brilleslijper-Kater, and Wolf 2010; Morton et al. 2020; Slesnick et al. 2009; 2023). Furthermore, like Altena, Brilleslijper-Kater, and Wolf (2010), this review also finds heterogeneity in program delivery and diversity in the level of description provided for interventions across articles. This is problematic, as it can be hard to decern whether articles are describing HF4Y or other types of programs that are similar (e.g., rapid rehousing and, supportive housing). In their recent review, Slesnick et al. (2023) argue that housing alone is not enough for youth and that supports play an integral role in their transition to adulthood. However, like the authors in the present review, they argue that supports and programs that work for youth are not clearly articulated and evaluated. The present review adds to the evidence-base on housing for YEH by examining HF4Y—a task that has not been done in isolation from other program models—and finds a need for more rigorous evaluation of HF4Y. Further, like Slesnick et al. (2023), the authors argue that more research on the establishment of good service and support-based interventions for youth should be prioritized.

Conclusion

HF4Y is a promising model that shows benefits for YEH. However, there is a dearth of evidence on the application of the model and the evidence base is not large or robust enough to conclude that HF4Y is an evidence-based solution to youth homelessness. The model principles articulated by Gaetz, Walter, and Story (2021) demonstrate a low-barrier, human rights-based approach to housing that is in alignment with the most important aspect—low-barrier access to housing—of Housing First for adults (Woodhall-Melnik and Dunn 2016). Like studies of Housing First programs for adults, it appears that fidelity to principles may be extremely important to the success of HF4Y programs (Gilmer 2016).

Additional research that speaks to the development and implementation of programs based on articulated HF4Y principles is critical to understanding and tweaking practices to provide the best possible outcomes for housing YEH. Hence, this review concludes that the literature on HF4Y is quite small, and contradictory outcomes on food security and housing security indicate a need for more research. In other words, the evidence base for HF4Y is growing, but still in its infancy. The principles behind HF4Y are promising; however, more work needs to be done on measuring program-specific outcomes before HF4Y can accurately be considered the gold standard for addressing youth homelessness.

Supplemental Material

sj-docx-1-uar-10.1177_10780874251314838 - Supplemental material for Housing First for Youth Who Experience Homelessness: A Systematic Review

Supplemental material, sj-docx-1-uar-10.1177_10780874251314838 for Housing First for Youth Who Experience Homelessness: A Systematic Review by Julia Woodhall-Melnik, Cassandra Monette, Chloe Reiser and Tobin LeBlanc Haley in Urban Affairs Review