Supportive Housing for Sexual and Gender Minority Individuals With Criminal Justice Histories: Challenges and Opportunities Identified by Providers and Clients

Abstract

Sexual and gender minority (SGM) individuals experience high rates of homelessness and criminal justice system involvement, underscoring the need for supportive housing services. To explore the service needs of this population, we interviewed providers (n = 11) and clients (n = 10) from eight supportive housing organizations working with SGM populations in Los Angeles County, California, USA. We used the Consolidated Framework for Implementation Research to synthesize interview responses into themes (by domain and cross-cutting). Takeaways included the need for investment in systems of care for vulnerable SGM populations; the particular marginalization of Trans individuals and providers that serve them; the roles of supportive housing staff, residents, and leadership in cultivating an affirming environment; the prevalence of discrimination and stigma within supportive housing programs and broader society; and the complex interrelationships among SGM identity, homelessness, and criminal justice system involvement. These findings have important implications for supportive housing services and related policy.

Keywords:

• sexual and gender minority
• LGBTQ
• homelessness
• criminal justice
• incarceration
• qualitative
• policy
• supportive housing

In the United States, sexual and gender minority (SGM) individuals disproportionately experience homelessness. SGM is an umbrella term referring to individuals with nonmajority sexual orientations (e.g., gay, lesbian, bisexual, pansexual, asexual) or gender identities (e.g., transgender, nonbinary, genderqueer, intersex); population estimates of these groups can be challenging to obtain, but collectively they account for about 5% of people in the United States (UCLA: The Williams Institute, 2019). Compared with the straight, cisgender majority, SGM adults are 2–4 times more likely to experience homelessness because of exclusion from housing access, economic marginalization, and greater physical and behavioral health problems (Badgett, Choi, & Wilson, 2019; Ecker, Aubry, & Sylvestre, 2019; Wilson, Choi et al., 2020). SGM youth are even more vulnerable, with up to 6 times greater rates of experiencing homelessness compared with non-SGM youth (largely driven by rejection from their families; Baams, Wilson, & Russell, 2019; Choi, Wilson, Shelton, & Gates, 2015). SGM people of color also disproportionally experience homelessness because of direct effects of racism, and the ways racism interacts with and compounds heterosexism and gender bias in SGM populations (Bruce, Stall, Fata, & Campbell, 2014; Choi et al., 2015; Wilson, Gomez, Sadat, Choi, & Badgett, 2020). Given these issues, equitable access to housing is an important policy goal for promoting SGM health.

Understanding the needs of SGM people experiencing homelessness requires consideration of the impact of criminal justice system involvement. Contact with the criminal justice system, such as arrest and incarceration in jails and prisons, is common among all people experiencing homelessness (Tejani, Rosenheck, Tsai, Kasprow, & McGuire, 2014; Tsai & Rosenheck, 2012). This is in part because poverty and homelessness are frequently criminalized, and in part because criminal justice system involvement leads to economic and social marginalization that increases risk of homelessness. Moreover, SGM people in general are involved in the criminal justice system up to 3 times as often as straight and cisgender people, again because their experiences of marginalization and poverty are criminalized (Grant et al., 2011; Hanssens, Moodie-Mills, Ritchie, Spade, & Vaid, 2014; Vera Institute of Justice, 2020). As with homelessness, experiences of criminal justice contact are especially prevalent among SGM youth and people of color (Choi et al., 2015; Hanssens et al., 2014; Meyer et al., 2017; Vera Institute of Justice, 2020; Wilson et al., 2017), representing the compounding effect of SGM identity on prevailing trends in arrest and incarceration. Overall, findings to date make clear that it is critical to address the multidirectional impacts of homelessness, criminal justice involvement, and SGM identity in housing policy and programs.

The present study explored the needs of SGM clients in the context of supportive housing interventions for people who have experienced homelessness and criminal justice system involvement. Although definitions of supportive housing vary widely (Tabol, Drebing, & Rosenheck, 2010), they generally share a combination of long-term, independent housing in community settings plus community-based services based on the client’s needs (e.g., case management, employment services, and mental health and/or substance use services). Sometimes, interim or transitional housing is provided prior to permanent accommodations (Hunter et al., 2021). Establishing stable housing is a critical need for many individuals transitioning out of homelessness or incarceration (with the latter group often being released without stable housing); for those who experience chronic homelessness, supportive housing programs can increase housing stability and reduce crisis care needs (Aubry et al., 2020; Gilmer, Manning, & Ettner, 2009; National Academies of Sciences, Engineering, & Medicine, 2018). The effects of supportive housing interventions on long-term health and service utilization outcomes are less clear, partly because of the limited research base and partly because of the serious health impacts of chronic homelessness (National Academies of Sciences, Engineering, & Medicine, 2018). The complexity of understanding long-term impacts is well illustrated by 6-year follow-up findings from the At Home/Chez Soi randomized trial of supportive housing services, which showed significant effects on housing stability but not health outcomes (quality of life, community functioning, substance use severity; Stergiopoulos et al., 2019) yet also found that receipt of supportive housing predicted participants’ experiencing better well-being trajectory profiles (Mejia-Lancheros et al., 2021). Nevertheless, the need to invest in supportive services for this population is clear (National Academies of Sciences, Engineering, & Medicine, 2018).

The effectiveness of supportive housing interventions depends on their ability to flexibly address the needs of different subpopulations. Given the high need for supportive housing interventions among SGM people who have experienced homelessness and criminal justice involvement, it is necessary to understand whether supportive housing programs are being implemented in ways that effectively meet the needs of SGM clients. Unfortunately, the importance of SGM identity is often overlooked within large-scale efforts focused on homelessness, especially when adults are the focus (Ecker et al., 2019), so strategies for maximizing the effectiveness of supportive housing interventions with SGM clients remain unclear. Evaluations of SGM-focused interventions can be found in the literature, but they are too rare and preliminary to draw strong conclusions; for example, Reback, Shoptaw, and Downing (2012) found that transgender-specific HIV Prevention Case Management reduced homelessness, but Nyamathi et al. (2016) did not find any effects on incarceration from a Nurse Case Management program focused on risky drug use among homeless SGM men. These findings suggest that SGM-specific tailoring may be an essential component of effective interventions for these populations, but provide limited information about how supportive housing programs might implement such tailoring. It is also unclear how such tailoring might best consider SGM clients’ prior experiences with arrest and incarceration, but research on supportive housing for individuals involved in the criminal justice system suggests effective programs need to provide more intensive supports and address justice system impacts on clients’ engagement and housing options (Aidala, McAllister, Yomogida, & Shubert, 2013; Fontaine, Gilchrist-Scott, & Horvath, 2011; Gillespie et al., 2017; Salem et al., 2015).

The current study had two primary goals: (a) capture provider and client perspectives on the facilitators and barriers associated with implementation of supportive housing services for SGM people with criminal justice histories, and (b) identify implications to improve practice and policy for these marginalized populations. Given the limited existing research available to guide our study, we took an exploratory approach to the research using qualitative methods. Qualitative research produces rich, detailed understanding of a phenomenon that can inform and/or complement more hypothesis-driven research testing specific solutions (Braun & Clarke, 2006; Camic, Rhodes, & Yardley, 2003). To ensure that our exploratory findings could inform policy and practice, we used the Consolidated Framework for Implementation Research (CFIR; Damschroder et al., 2009) to organize our qualitative results. CFIR is the most widely used framework for implementation research, providing a comprehensive overview of how a program’s implementation is influenced by characteristics in five major domains. More specifically, using CFIR allowed us to identify the needs and experiences of SGM supportive housing clients with criminal justice histories relating to characteristics of (a) the intervention (i.e., supportive housing services), (b) the individuals involved (i.e., supportive housing providers and clients), (c) intraorganizational contexts or the inner setting (i.e., organizations delivering supportive housing services), (d) interorganizational social and policy contexts or the outer setting (i.e., county, state, and national policy and practice context), and (e) the implementation process for supportive housing services.

Materials and Methods

Study Context

This study was based in Los Angeles County, California, which has the largest unsheltered population in the United States. The latest estimates indicate that more than 66,000 people in the county experience homelessness at a given point in time (Los Angeles Homeless Services Authority, 2020). The county also operates the nation’s largest jail system—averaging 17,000 daily detentions (Los Angeles County Sheriff, 2018). Within this context, Los Angeles County has invested in various supportive housing programs aimed at improving housing stability among vulnerable groups, providing a wealth of experiences to draw upon for our sampling and qualitative data collection.

Participants

Study participants were providers (n = 11) and clients (n = 10) from supportive housing programs located within Los Angeles County. Given the small size of the sample, we provide limited demographic descriptions to protect participant confidentiality. The 11 providers we interviewed represented eight supportive housing programs with experience serving SGM clients; we conducted a group interview with three providers from one program, and separately interviewed two providers from another program. More than half (five) of the programs offered SGM-specific services, one of which was specifically for Trans clients; the other three reported serving large numbers of SGM clients but did not offer population-specific services. Two of the programs offered traditional supportive housing programs, another two did not manage the housing but worked closely with the client to obtain housing and supportive services, and the remaining four offered Recovery Bridge Housing (i.e., time-limited, abstinence-focused peer-supported housing coupled with outpatient substance use treatment). Interviewees’ roles at their agencies varied, but generally involved a combination of program administration and direct client services.

The SGM supportive housing program clients we interviewed represented a range of sexual orientations (four gay men, four bisexual men and women, one pansexual individual, and one who did not disclose) and included seven cisgender and three transgender individuals (all Trans women of varied sexual orientations). We did not identify any lesbian women, Trans men, or nonbinary, asexual, or intersex individuals to interview. Half of the clients were living in interim supportive housing, another three were in Recovery Bridge Housing, and two were in permanent supportive housing placements. We did not collect details about clients’ criminal justice histories because we expected that would be a barrier to recruitment. We also did not systematically collect information about participant race or ethnicity, but most clients self-disclosed identifying as white, Black, or Hispanic during their interviews. We offered interviews in both English and Spanish; one client completed an interview in Spanish and the rest were in English.

Procedures

The Institutional Review Board at RAND Corporation approved all recruitment and data collection procedures for the present study. Recruitment began in October 2020, and all activities were completed virtually because of the ongoing COVID-19 pandemic.

Recruitment

We used a phased approach to recruitment that made use of snowball sampling, in which study participants recommended other individuals who could participate (Kemper, Stringfield, & Teddlie, 2003). Provider recruitment was conducted via email. To be eligible for an interview, providers needed to work at an organization that provided supportive housing interventions for people experiencing homelessness, and self-identify as being able to speak in depth about clients’ needs related to SGM identity and criminal justice history based on their professional experience. We began by soliciting recommendations for organizations to interview based on existing relationships with Los Angeles County-contracted supportive housing providers (Brooks Holliday, Hunter, Dopp, Chamberlin, & Iguchi, 2020); the first author sent email invitations to nine recommended individuals representing five organizations, and scheduled interviews. After each provider interview, we collected recommendations for additional individuals to contact, which produced another 11 recommendations representing five additional organizations. Thus, we sent 20 invitations to 10 organizations, of which one self-identified their organization as ineligible, two declined, and six did not respond. We ultimately completed 11 interviews with eight organizations.

At the end of each provider interview, we also requested that providers share information about our study with clients who might be eligible for an interview. We provided informational flyers to these providers that gave a brief description of the study for clients and contained our contact information. We also shared with providers consent-to-contact forms that a provider could fill out with a client and send to the research team (who would then contact the client for eligibility screening, study description, and consent). All client-facing recruitment materials were offered in both English and Spanish; the third author translated the Spanish materials. Overall, we received three consent-to-contact forms and were contacted directly by 17 additional clients; these clients came from four of the eight organizations for which we interviewed providers. To determine client eligibility, telephone screening was completed by the second author (English-speaking) or third author (Spanish-speaking), who then scheduled interviews with eligible individuals. Eligibility was based on the client reporting an SGM identity, currently receiving supportive housing services, and having been housed in a criminal justice setting during adulthood (defined as detention in jail or prison for at least one day). Of the 20 clients who made contact in some way, we screened 15 of them; with the others, we were unable to make contact and complete the screening, in one case because the client ended the call when asked about their criminal justice history. After screening, 13 clients were found to be eligible (the other two were not currently in supportive housing), and 10 of those eligible scheduled and attended an interview. We did not engage in formal snowball sampling efforts with clients, but we did encourage them to share the study flyer with peers who might be eligible.

Data Collection

Data collection lasted about 3 months, from November 2020 to February 2021. Interviews were completed via a secure Microsoft Teams meeting, which offers video- and audio-conferencing options; in-person interviews were not possible at that time because of the COVID-19 pandemic. Providers all used the video-conferencing option. Many clients did not have access to technology for a video-conference meeting, so we gave them the phone number to join the meeting, and sometimes called clients directly from the Teams meeting in cases where that was more convenient.

All English-language interviews were led by either the first or second author, with the other author taking detailed notes on the conversation; the third author led the Spanish-language interview using a translated interview guide while another bilingual individual (listed in the Acknowledgments) took notes in English. In all cases, the interviewer followed a semistructured interview guide to elicit information about participants’ experiences with supportive housing for SGM individuals with criminal justice histories. English copies of the provider and client guides are available in the Appendix.

The provider interview asked about supportive housing services provided at the organization; housing needs for SGM clients (including those with incarceration histories), and program and system capacity to meet those needs; barriers and facilitators to providing supportive housing services for SGM clients (including those with incarceration histories); and the future outlook of their services.

Topics covered in the client interview included supportive housing services the client is currently receiving (and has received in the past); needs and desires for supportive housing; experiences in supportive housing; and challenges and benefits related to supportive housing. For all these topics, the interviewer probed how the client’s SGM identities and criminal justice history related to their perspectives.

Generally, provider interviews lasted about 45–60 min and client interviews lasted about 30–45 min. When permitted by the participant, we audio-recorded the interviews to allow for verification of notes and verbatim transcription of illustrative quotes; all providers and all clients but one permitted audio-recording. Participants’ study materials (i.e., audio recordings, notes) were all identified using a unique, anonymous participant identification number to maximize confidentiality; the audio recordings were destroyed once transcription and analysis were complete. Providers did not receive compensation for the interviews, which was explained in the consent process. Clients received a $25 gift card in exchange for their time; we offered various electronic and physical gift card options, and the client chose their preferred option after the interview.

Data Analysis

After the completion of all interviews, the first and second author completed a conventional content analysis (Hsieh & Shannon, 2005) of the interview notes. First, we reviewed the notes and developed a codebook of barriers and facilitators to supportive housing interventions for SGM individuals with criminal justice histories, using a combination of deductive and inductive coding. We inductively analyzed the data by identifying emergent codes from participants’ responses, then organized the identified codes within the five domains of CFIR (Damschroder et al., 2009). The codes represented (a) general themes that summarized the content of participants’ responses (including not only cohesive responses, but also inconsistencies—i.e., negative case analysis) and (b) more nuanced subthemes that focused on a specific aspect of a general theme. We grouped identified themes within the five CFIR domains, and further categorized each theme and subtheme as either a barrier that impeded supportive housing implementation or a facilitator that promoted supportive housing implementation. We also compared responses between provider and client interviews, and because we found substantial overlap, we chose to develop a combined codebook but also noted which codes were derived from provider interviews, from client interviews, or both. Once the coding by CFIR domain was complete, we also identified cross-cutting themes that were present within a number of specific themes (usually from multiple CFIR domains) as a way to inform policy and practice.

The first author coded all interview notes with the initial codebook and identified cross-cutting themes, and the second author reviewed all coding to ensure agreement. When disagreements were identified, the first and second author discussed them until they achieved consensus. Once the coding was finalized, all authors reviewed the themes and subthemes, and agreed that the interviews had reached saturation (i.e., additional interviews were unlikely to result in meaningful changes to the codebook) for sample-wide themes. Finally, we developed written summaries of all themes (by domain and cross-cutting), and we selected and transcribed exemplar quotes that illustrated themes and subthemes; note that the cross-cutting themes were too complex to capture in a single quote.

Researcher Reflexivity

Qualitative methods emphasize how researchers’ identities and experiences influence the ways they collect, interpret, and present data—a process known as reflexivity (Watt, 2015). In this study, the lead researchers (first and second author) are white gay men. The other members of the research team identify as a Latina heterosexual woman and a white heterosexual woman, respectively. All team members were cisgender, so we sought consultation with colleagues of Trans experience to inform our data collection and analysis approach. No member of the research team had a history of lived experience with homelessness or criminal justice involvement, but all had extensive professional experience working with service providers who served these populations and service recipients who have such experiences. Our team viewed this project as an expression of allyship to SGM individuals and communities and to all groups experiencing disproportionate homelessness and criminal justice involvement—e.g., our allyship to Black individuals and communities is also relevant. We aimed to collect and analyze data that reflected SGM clients’ experiences and to present results in a way that would be useful for those clients and the agencies working with them—but recognize that our own experiences may still have colored our interpretations.

Results

We begin by summarizing the themes for each CFIR domain. For each domain, we provide a narrative description of themes and subthemes (the themes are denoted in italics) and one exemplar quote. Accompanying tables provide a detailed list of themes and subthemes, noting the type (facilitator or barrier) and reporter (providers, clients, or both), plus exemplar quotes. Note that in quotes, interviewees used variations of the term LGBTQ+ (lesbian, gay, bisexual, transgender, and other queer identities) rather than SGM. We also identified five cross-cutting themes that spanned the CFIR domains, and these are described after the individual themes. The order of presentation of material within the text and tables was chosen to provide a logical flow of information; it is not meant to convey information about salience or prevalence.

Intervention Characteristics

Table 1 details themes related to the characteristics of supportive housing programs. Overall, participants described the need for supportive housing interventions to be comprehensive—with specific components that address clients’ interrelated needs, including those related to their SGM identity and criminal justice history—but with housing as the central component for promoting long-term outcomes. As one provider put it,

We have to be willing to address head-on mental health and substance use, and…for me, it’s basic [SGM cultural competence] training stuff, we’re not gonna get the outcomes because the people aren’t gonna feel respected…[such as] understanding the continuum of LGBTQ+, the willingness to meet people where they are so that you can be supportive of the individual to then find the housing that will work for them.

Table 1. Intervention characteristic themes about supportive housing for sexual and gender minorities with criminal justice histories.

Theme/subtheme	Type	Reported by		Exemplar quote
		Providers	Clients
Provides stability for long-term goals	Facilitator	X	X	“On the street you’re not really worrying about, like, therapy or going to, like, counseling or working with your [Intensive Case Management Services] worker. You’re just worried about survival. But when you have a stable place to live then you can actually…your hierarchy of needs is met and you can worry about other things and other lists and tasks.” (Client)
Comprehensive care models with flexible timelines	Facilitator	X	X	“They have a team for me, so I have my case manager, my substance abuse counselor, my therapist, an occupational therapist…and any time I need help with filling out [Social Security] papers, I can always ask one of the case managers.” (Client)
Priority of behavioral health services	Facilitator	X	X	“It’s an array of different needs, but I think that in order for us to get them stable, we start with providing them mental health [services]. And then, we start working with the rest of the pieces of the puzzle. At the end of the day, we want all of our clients to have a successful exit, right?” (Provider)
Clients may have different priorities than behavioral health services	Barrier	X	X	“They do what they can and, you know, it’s, like, to help you find a place to live, permanently. So that’s the goal. So in the meantime…they want us to go to all the groups and stuff like that. Like, if you have depression…most of the times, I don’t feel like I need that group. Because, I feel like, me personally as myself, I already surpassed that.” (Client)
Focus on population-specific issues	Facilitator	X	X	“I’ve gotten feedback from some of our fellows—they went to apply at an agency that is there to help them, and when they identify themselves as LGBTQ, they felt like they were pushed to the back of the line. So a lot of them don’t want to be living their truth when it comes to try to apply for housing or whatever because they’re afraid they’re going to be discriminated against. Being able to come and sit with someone and be your authentic self and still get help is kind of what we’re here for.” (Provider)
Offering criminal justice outreach and reentry programs	Facilitator	X	X	“Folks write us from prison or a detention center. We connect with them, we have some people that actually write them, and then when they get out they actually just come here, and we help connect them with other services. … Their parole officers—all my clients, I get the same response from them—is that their parole officer tells them ‘I don’t know where to place you’ because they’re Trans.” (Provider)
Different views on appropriate program expectations	Barrier	X	X	“We are not aligned with the [district attorney], or the judges, or the court system. We advocate for our clients. And when someone relapses, they are not discharged. The consequences are much more subtle. We don’t feel that you can discharge a patient for exhibiting the symptoms of the disease. Particularly if you’re in [Recovery Bridge Housing] where we’ve already acknowledged that they’re homeless.” (Provider)

Note. LGBTQ= lesbian, gay, bisexual, transgender, and other queer identities.

Two of the intervention themes concerned supportive housing program characteristics that were viewed as helpful generally, not just for SGM clients with criminal justice histories. In one theme, participants described how supportive housing programs provide stability for clients to achieve long-term goals. Homelessness and incarceration are highly disruptive to people’s lives, and many do not experience the safety and security needed to plan for the future until they have obtained stable housing. Both interim and permanent supportive housing were seen as accomplishing this purpose, although permanent housing was an important goal for many interim housing clients. In the other theme, participants viewed the comprehensive care model used by supportive housing programs as beneficial. It was described to us that the comprehensive care model is needed to address clients’ interrelated medical needs (i.e., HIV, Hepatitis C, disability, Trans-affirming hormones/surgery), behavioral health needs, income/employment needs, and other challenges within a common framework—often facilitated by a case manager. Participants noted that flexible timelines for services are helpful to ensure that all client needs can be addressed; this can be a challenge in time-limited services like Recovery Bridge Housing, further underscoring the importance of flexibility.

Focusing now on SGM clients with criminal justice histories, participants identified behavioral health needs (e.g., mental health and substance use treatment) as being particularly important to address. Providers viewed behavioral health services as a high priority for this population, and indeed some clients described the mental health or substance use treatment components of their supportive housing as being very helpful. In addition, participants identified the importance of incorporating topics related to SGM identity and criminal justice history into the behavioral health services; examples include identity development, trust and belonging, family reunification, and the role of sexuality in relationships. However, participants also acknowledged that clients may be focused on different goals—such as stable housing or employment—and less interested in behavioral health services. Several clients indicated that they did not find the behavioral health component of their supportive housing program to be useful or necessary.

The remaining themes related to supportive housing program components for clients with criminal justice histories. Participants noted the importance of having formal criminal justice outreach and reentry components within the supportive housing program; these services often begin when clients are still incarcerated and assist with the transition back into the community. Without such components, it can be difficult for supportive housing programs to identify and engage clients with criminal justice histories successfully. Finally, we noted providers’ differing views about the expectations that supportive housing programs should have around client behavior. SGM clients with criminal justice histories may exhibit challenging/harmful interpersonal behaviors or may engage in substance use in sober living spaces. Some supportive housing programs work closely with the criminal justice system, and clients may lose their supportive housing or be reincarcerated after engaging in such behaviors; other providers advocated for more flexible expectations, noting that these behaviors often relate to clients’ behavioral health challenges (e.g., trauma, substance use). Clients consistently expressed a preference for the latter approach.

Individual Characteristics

Table 2 details themes related to the characteristics of the individuals (providers and clients) involved in supportive housing programs. Generally, it was seen as critical that program staff be able to relate to clients whose backgrounds (SGM identities and criminal justice experiences) predisposed them to feel rejected, judged, and even endangered by untrusted others. One client described how

When I first came here six months ago, I felt scared, because all these thoughts and feelings were coming back to me about other [supportive housing] places that I’d been to…I was totally wrong about it, I got here and it was like, they accepted me with open arms, they were like we’re here for you…I thought they were full of shit, but I started seeing, like oh my god these people actually do care, they really do.

Table 2. Individual characteristic themes about supportive housing for sexual and gender minorities with criminal justice histories.

Theme/subtheme	Type	Reported by		Exemplar quote
		Providers	Clients
Staff who can relate to clients	Facilitator	X	X	“We are unapologetically LGBTQ+, and I think that that serves us well…. The [clients] know that ‘we are them.’ At least half of our staff is LGBTQ, so when they walk through the door, they’re seeing people where they may feel more comfortable, that they identify with. So just being who we are, we are a beacon for our LGBTQ+ community.” (Provider)
Staff limitations	Barrier	X	X	“I believe that in order not to have these kind of experiences, it would be help[ful] to know of companies and people who understand and accept [transgender people]. It’s like [case manager] might say ‘I’m going to send you over to this company’…but the company has no familiarity with LGBTQ. (Client)
Varied cultural sensitivity across service providers	Barrier	X	X	“We’ve created a small network of culturally competent LGBT substance use and co-occurring providers. That network is about three providers right now. There are lots of agencies out there that say ‘Oh yeah, we treat ‘em!,’ you know, like that’s something significant, but they have little or no competency…and we had to go through our own personal journey to get there [with our competency].” (Provider)
Limited client self-disclosure to providers	Barrier		X	“I never told [my case manager] about the names that they were calling me. I just figured that people just anywhere wouldn’t get over the fact that I was hanging out with a [transgender woman]…I think if I would have told them they would have done something about it, but I just never brought it up for some reason. I just figured it was part of life.” (Client)
Diversity of people within SGM populations	Facilitator	X	X	“The majority of our applicants are Black. We believe that, because of the disproportionality of our numbers across the board, that this is our opportunity to apply some equity to our issues. Not to say that others are not experiencing [issues with homelessness, mental health, and addiction], but our numbers are so out of whack that at some point it is time for us to have some special attention.” (Provider)

Note. LGBTQ+ = lesbian, gay, bisexual, transgender, and other queer identities (some interviewees used the variations LGBTQ or LGBT).

Unfortunately, the lack of such staff was seen as a key limitation of many supportive housing programs.

Both providers and clients felt it was very important that supportive housing programs employ staff who can relate to SGM clients with criminal justice histories and are committed to helping those groups. Employing staff with lived experience can be helpful—for example, one provider shared “I know the program works because I’m proof”—but most interviewees did not think lived experience was required of every staff person. Rather, examples of the characteristics highlighted include being supportive, encouraging, understanding, knowledgeable, open-minded, and able to make clients comfortable. However, participants also acknowledged that such staff can be difficult to find within social services; they described many limitations of staff in terms of relevant knowledge, skills, and experiences. This included both staff limitations related to SGM- and criminal justice-specific topics and more general limitations in their role as a supportive housing provider.

In addition, providers and clients acknowledged varying levels of cultural sensitivity among the various providers involved in supportive housing services, with many agencies and individual service providers in the community seen as unprepared to work with SGM clients or clients with criminal justice histories. Thus, even if the supportive housing staff can relate to this population, it can be a major challenge to ensure that all components of the comprehensive supportive housing services (e.g., mental health, substance use, medical, case management) are sensitive to their needs. Participants generally identified individual providers, rather than entire organizations, as having sufficient competence. Perhaps as a result of this dynamic, client interviewees often described being reluctant to disclose their SGM identity or criminal justice history to their service providers. These clients reported believing that certain providers would be supportive or at least nondiscriminatory (e.g., “I don’t think it would matter to them”), but still choosing not to disclose; the same clients also reported experiences of discrimination with other providers or settings (detailed in later themes). Our interpretation is that these clients felt it was safer to not disclose with providers as a general rule, and then made rare exceptions for those individuals who clearly demonstrated sensitivity to the client’s identities and experiences.

Finally, providers and clients identified many important dimensions of identity, experience, and diversity within SGM communities that supportive housing programs need to consider. Criminal justice history is certainly one such dimension, and participants recognized the diversity of sexual orientations and gender identities within SGM communities as well. Other examples noted by participants include racial and ethnic identities, language, immigration status, disability status, and socioeconomic status. Supportive housing programs that are SGM-friendly may be more or less inclusive of individual clients across these other dimensions.

Inner Setting Characteristics

Table 3 details themes related to the characteristics of the organizations delivering the supportive housing interventions. Both providers and clients described the importance of the housing program’s interpersonal climate—and contrasted their experiences with affirming versus discriminatory programs—while also contributing unique perspectives on characteristics of well-functioning organizations. A provider summed this up by saying,

With the culture of our houses, we make sure we are sensitive to whatever traumas, whatever stigma the reentry clients come across. We make it a point to make sure it doesn’t live in our spaces, so they feel totally comfortable, totally accepted, not judged, and I think that’s integral to restoring the resiliency in the clients before they go back into the community. They wanna be strong, resilient, confident, a thriving member of the community.

Table 3. Inner setting characteristic themes about supportive housing for sexual and gender minorities with criminal justice histories.

Theme/subtheme	Type	Reported by		Exemplar quote
		Providers	Clients
Affirming environment	Facilitator	X	X	“I found this [supportive housing] company, which acts like more than just your friends. They act like a community of family members. They have found you a place where you can go in and let your dust settle. And you can, you know, let things kind of clear in your head and give you space to do what you need to do. And they’ll walk with you, if necessary, for as long as you feel it’s necessary.” (Client)
SGM-friendly as a proxy for affirming	Facilitator	X	X	“Where I’m going to be accepted for who I am no matter what that is. That’s what I was looking for. It didn’t matter to me so much that it be LGBTQ, as long as it was an environment where they’re accepting of people of any sexual preference, religion, race, whatever. But that tends to be more inclusive in the LGBTQ community. So I guess that’s why I was hoping for that, and would seek that out again. It’s kind of a crap shoot otherwise.” (Client)
Leadership support	Facilitator	X		“Working at an organization whose upper management supports that community and supports the needs of that community. And the amount of training that the organization does around cultural sensitivity and gay and lesbian and transgender issues. Every year there’s just a ton of mandated trainings that we have to go to, which I think results in having a team that can work better with that population.” (Provider)
Discrimination in supportive housing programs	Barrier	X	X	“When I got out of prison, I was looking for housing—you know, [Recovery Bridge Housing] and stuff like that because I wanted the help. And I didn’t find any LGBT-friendly places because there wasn’t any, or if there was I never found one. … For me being a gay man, it was very hard because I couldn’t be myself first of all, in the place where I was staying. It was like I was leading a double life.” (Client)
Team communication	Facilitator	X		“Just having the ability to have my navigators—they have a relationship with the patient’s doctors. Of course things are confidential, we need consent to release…but just being able to have that contact. We’ve got doctors and [licensed vocational nurses] coming in to our case conferences, when we talk about cases and work on linking clients to different services.” (Provider)
Being visible and accessible within the community	Facilitator	X		“We also are one of the few programs that, on a Friday afternoon, or the day before Thanksgiving, or Christmas Eve, we do take people in. There’s not a time where we shut down and people aren’t in the office and we’re not answering the phone. And we don’t have a machine or a system, it’s live people all the time. So those barriers are eliminated.” (Provider)
Conflict between program rules and client expectations	Barrier		X	“Anything I would change would have nothing to do with it being LGBTQ. It would be more like, maybe a little more diligence about holding people accountable for cleaning the house and stuff like that. Which has nothing to do with sexual preference or identity.” (Client)

Note. LGBTQ= lesbian, gay, bisexual, transgender, and other queer identities (some interviewees used the variation LGBT).

This illustrates how trustworthiness must extend to the organization level, not just individual staff.

Both providers and clients highlighted the importance of an affirming environment within the supportive housing program, meaning that clients are allowed the time and space to be their authentic selves without judgment. Interviewees noted that both staff and clients contribute to the environment being affirming, and that all individuals feel they are safe to be themselves in the environment—including with their SGM identities and experiences with the criminal justice system. An important distinction is that many providers and clients desired an environment that was affirming generally, not affirming for SGM people specifically, but stated that they viewed SGM-affirming environments as a good proxy. For example, one provider described how clients had benefitted from integrating their SGM-focused group housing with other housing, while taking steps to ensure the housing environment remained inclusive. Another related point raised by providers was the role that their organizational leadership played in creating an affirming environment. Interviewees described a variety of leadership roles (e.g., agency administrators, advisory boards) and mechanisms (e.g., communicating a vision, creating structures, obtaining client feedback) by which organizational leaders created an affirming environment.

In contrast to the ideal of an affirming environment, many (but not all) clients reported experiencing various forms of discrimination within their supportive housing services related to LGBTQ+ identity and criminal justice history. This discrimination ranged in severity from social exclusion and comments to repeated harassment, to at times violence, and was perpetrated by other clients and by housing program staff. For example, one female client described being raped by a housing staff member after disclosing her bisexual identity to him. The providers interviewed also acknowledged the prevalence of discrimination encountered by SGM clients with criminal justice histories within supportive housing services, although their accounts were less specific than the clients’. The clients generally identified the most harmful discrimination as taking place in prior placements (including jail) rather than placements in the supportive housing programs from which we had interviewed providers.

From the provider interviews, we also identified two themes related to organizational functioning that providers felt promoted their success. One theme was team communication and care coordination, which related to the complexity of services involved in supportive housing programs. Some providers described how their agencies had well-developed communication practices that helped all team members stay informed and work together. The other theme came from organizations that are particularly visible and accessible within the community, which the provider interviewees viewed as promoting client engagement. Organizations that have a strong presence in the community, contribute to efforts that benefit the community, and can be easily accessed by clients (e.g., have convenient hours) can more easily earn the trust of clients.

Finally, many clients living in interim or other shared housing situations described having concerns or disagreements about program rules with staff or other clients. Unlike the earlier theme about discrimination, these concerns and disagreements reflected their general expectations about living conditions. Examples of client concerns included rules (and enforcement thereof) around cleanliness, food and cooking, sharing rooms, and curfews. One client describe feeling treated “like a criminal” in a previous supportive housing program because of the rules, but overall, these concerns illustrate that clients’ SGM identities and criminal justice histories are not always central or high priority within their experiences of supportive housing.

Outer Setting Characteristics

Table 4 details themes about the characteristics of the social and policy contexts external to the supportive housing programs, namely in Los Angeles County in the U.S. state of California. Overall, supportive housing clients and the organizations serving them experience major structural barriers rooted in systematic, interconnected marginalization of people with SGM identities, criminal justice histories, and other factors contributing to homelessness—yet have found ways to succeed within these structures. These barriers result in multiple pathways back to homelessness, with one client stating “I got denied [an apartment] because of a debt that I have due to identity theft…[looking for housing is] like having a job while you’re also looking for a job.” Another client noted:

Since I have been 18, I had to look for work, and I had times when I was ashamed and treated badly, and felt like they don’t want to give me work…I need to find a job and I am scared to apply because I am transgender.

Table 4. Outer setting characteristic themes about supportive housing for sexual and gender minorities with criminal justice histories.

Theme/subtheme	Type	Reported by		Exemplar quote
		Providers	Clients
Discrimination in society	Barrier	X	X	“I’m so protective that when someone applies for housing, for their own apartment, and they don’t get, I go and want to find out. Like, let’s have somebody else submit an application and see what happens. I have this one staff member…she’ll go apply for an apartment after somebody gets turned down. And she’ll get it.” (Provider).
System-of-care issues	Barrier	X	X	“It is really hard when you are told that by law you can only have one case manager that is helping you with housing when that case manager is overwhelmed…there needs to be more social workers or case managers and to be able to have time, because if I just rely on one person and that person has all these other cases, it is really hard to focus on me!” (Client)
Use of informal networks	Facilitator	X		“I know the buildings and the landlords to say to a Trans person, ‘These landlords are really cool. Let’s look and see if they have apartments for rent.’ It’s an inside game.” (Provider)
Alternative systems of care	Facilitator	X	X	“I’m under the mental health umbrella, so [mental health services] go along with me. And, it’s actually a feather in my cap because of my situation. It gives you, I think, some built-in protections.” (Client)
Cycles of marginalization	Barrier	X	X	“Incarceration is kind of just, I guess, a byproduct of their life circumstances. It’s just what they do for survival, being sex workers, or stealing, or drug addiction because of inability to deal with life situations. … Any time you don’t have a source of income to be able to take care of yourself, you gonna survive by any means necessary…. For those that are sex workers, they have to look a certain way, they have to be appealing. So of course they gonna go to a store and they gonna shoplift, they gonna do stuff to keep up their appearance.” (Provider)
Criminal justice system impact on SGM identities	Barrier	X	X	“I would hear people say ‘Is that guy really gay?’ … I guess I don’t act or look a certain way, you know what I mean? It was very hard for me, I didn’t feel like I fit in…. But then as the counselors and the other staff were like ‘Yeah, you’re gay, just be yourself and don’t try to impress nobody else’…. In prison, if you’re gay, it’s a no-no, keep it on the hush-hush, on the downlow ‘cause will get your ass kicked, basically…I had to act and be a certain way to be accepted by the other guys…for 15 years, I was taught to be a certain way.” (Client)
Stigma	Barrier	X	X	“For [my parents] it was acceptable for many years for me to be gay and that is all…my father I don’t believe accepts me as transgender—my mother yes, my father no because he says to me, ‘If you want to be gay, just act like the other men, no one has to know your business.’ So there are members of my family who accept it and other members who don’t accept or don’t have the right information.” (Client)
Formal policy barriers	Barrier	X	X	“If you’ve transitioned, and you’ve gotten your name changed from when you’re in between treatment and sober living, there’s no history for you. If somebody runs a credit check, it’s not that you have bad credit, you don’t have anything. So, if we keep them in sober living for over a year, maybe a year and a half, we can then establish that yes they’ve paid rent here, and they can start to have some credit.” (Provider)
Location of housing embodies stigma	Barrier	X	X	“They seem to have an area, a door, an entrance to where they want to place [supportive housing]. And I noticed that…everybody, for whatever reason, we kind of converged in the space because of, lack of a better term, being displaced. And some displacements could have been much more rockier than others, and much more severe. … That kind of plays out, and you see that a lot.” (Client)

Participants described a range of discrimination experiences within society at large for SGM clients with criminal justice histories. Participants often observed such discrimination in the context of seeking permanent housing, jobs, or social services; clients frequently have their applications rejected (e.g., for an apartment or a job) or otherwise feel excluded when trying to navigate these spaces. These experiences sometimes lead clients to hide their identities or experiences while trying to reach a particular goal, and to only live more openly once they have secured housing, work, or services. Clients described fewer experiences of violence within the outer setting than they did within supportive housing services, but violence still occurred; for example, one client described an instance in which a man was attacking transgender sex workers in an area of downtown Los Angeles where she was engaging in sex work to support herself.

There was a strong consensus among participants that there are many issues with the local system of care for SGM supportive housing clients with criminal justice histories. The system of care for SGM-affirming supportive housing is small, underresourced, and poorly coordinated, with inconsistent quality of services, and these issues are compounded for clients who have criminal justice histories. Nevertheless, two subthemes indicated ways in which providers and clients had found ways to navigate within these system of care issues. One subtheme, described only by providers, was the use of informal networks to coordinate among partner agencies for supportive housing and related services. Providers described how they build relationships with service agencies, landlords, and other individuals who have SGM identities or are SGM-friendly; they then assist their clients in connecting with those individuals and agencies. The other subtheme was the presence of other systems of care that were better resourced and, although not specifically targeting SGM populations, included high proportions of those groups. Examples included the supportive housing systems for people living with HIV and for people with substance use disorders (e.g., Recovery Bridge Housing).

Participants also articulated the complex interconnections among SGM peoples’ experiences of marginalization within society (e.g., difficulty finding housing or jobs), poverty, and crime. For many SGM people, their involvement in the criminal justice system begins with crimes of survival within poverty and homelessness (e.g., stealing, sex work) or with behavioral health challenges resulting from discrimination and trauma (e.g., illegal substance use, interpersonal violence). Criminal justice history only furthers their marginalization by creating additional barriers to housing, employment, and behavioral health recovery—thus fueling a cycle of continued illegal behavior, arrests, and incarceration. As one provider put it, “They come out to society, and nothing is here for them.”

Another aspect of these complex interconnections is the theme of how experiences in the criminal justice system impact SGM identity and expression. Participants described numerous ways in which jails, prisons, courts, and probation offices created unsafe and nonaffirming environments for SGM people, leading them to hide their identities and try to pass as straight and cisgender—or else risk violence and unfair treatment by those with power in the system.

Finally, it is important to recognize that societal stigma against SGM people and people with criminal justice histories underlies most of the barriers identified in this analysis. Clients who have these identities and experiences are often viewed as abnormal, dangerous, or otherwise problematic and as undesirable neighbors, employees, clients, or community members. Even within SGM communities, people may express the same types of stigma—especially toward those with a different SGM identity and those with criminal justice histories. Most of the stigma described was informal, based on commonly held attitudes or prejudices within society, but sometimes stigma was identified within formal policy barriers against certain groups. In particular, Trans people may be barred from a variety of services because of issues with documentation (e.g., name or gender not matching official documents) or requirements that services are based on sex assigned at birth (e.g., group shelter or housing arrangements), and people with criminal justice histories may be denied housing or employment opportunities based on background check results. Finally, the physical location of supportive housing options can embody stigma, as providers and clients perceive that this housing is often only permitted in undesirable or underresourced communities. This increases exposure to crime, substance use, and other challenges while reinforcing clients’ marginalization.

Process Characteristics

We identified only one theme related to the process of implementing supportive housing interventions for SGM clients with criminal justice histories. This theme was a facilitator, and appeared only in the provider interviews. Specifically, providers described ongoing efforts to expand supportive housing services for SGM clients (generally and for those with criminal justice histories). Many providers expressed a hopeful outlook about their ability to serve more clients in the future, and to expand services for groups that are currently the most underserved (e.g., Trans clients). For example, one provider stated:

We are aggressively identifying demographics that have housing needs and we’re doing our best to rise to the occasion…. It’s kind of a replicable model, almost like a machine: identify the demographic, get the funding, build the program, service the demographic. So it’s a well-oiled machine and it’s definitely a machine that’s growing and it’s getting more sophisticated.

Some providers acknowledged that it can be challenging to secure enough resources (e.g., hiring, training, space) to support growth, but this was rare, and even these providers still generally described their growth as a facilitator.

Cross-Cutting Themes

Based on our synthesis of the key themes and type of themes (i.e., facilitators and barriers) expressed by the providers and clients we interviewed, we developed five cross-cutting themes that capture implications for policy and practice regarding supportive housing for SGM with criminal justice histories.

There Is an Urgent Need to Increase Investment in Comprehensive Services That Can Support SGM People Who Have Experienced Homelessness and Criminal Justice Involvement

Participants detailed numerous ways in which the resources available in Los Angeles County are insufficient to meet the needs of this population. This implication most closely relates to the outer setting theme about system-of-care issues, but cut across the other CFIR domains: the need for comprehensive supportive housing programs that can be tailored to clients’ needs and goals (e.g., behavioral health services, criminal justice outreach and reentry: intervention domain), and for working with supportive staff in an affirming environment (individual and inner setting domains, respectively). At present, many supportive housing placements do not offer such services but instead were described as sites of harassment, violence, and discrimination (inner and outer setting). However, providers described some hopefulness about their ability to continue expanding services for this population (process domain).

Trans and Nonbinary Clients (and the Agencies That Serve Them) Are Particularly Marginalized and Need Additional, Targeted Supports

Within the inner and outer setting themes, participants frequently identified transgender people as having more frequent and intense experiences of discrimination and stigma compared with lesbian, gay, or bisexual people. Even among Trans individuals, participants described additional marginalization of Trans men, nonbinary individuals, and Trans people of color. These groups were described as experiencing discrimination even within SGM communities, and they experience formal policy barriers that other SGM groups do not (e.g., exclusion from gender-segregated settings, barriers related to legal identification). Participants also highlighted these issues when describing the importance of recognizing clients’ diverse characteristics and needs (individual domain) and of services addressing population-specific issues (intervention domain), such as assisting with changing the name and gender on legal documents. Again, there was strong interest among providers in expanding Trans-specific services (process domain).

Careful Selection and Training of Staff, Residents, and Leadership Is Critical to Cultivating an Affirming Environment

Participants identified specific features of an affirming supportive housing program environment (inner setting domain) that overlapped with the characteristics of individual staff and clients (individual domain) as well as the organization’s leadership structure, processes, and vision (inner setting). The program’s core components, partnerships, and approach to clients who violate program expectations (intervention domain) also codify the affirming environment; for example, one provider described how their organization will actively exclude potential clients from placement if they are unwilling to live with SGM people, as including such clients would undermine inclusiveness.

Supportive Housing Providers Must Be Prepared to Help SGM Clients With Criminal Justice Histories Navigate Societal Discrimination

Participants described the most effective supportive housing providers as taking an active, collaborative role with clients in helping to mitigate the impacts of societal discrimination. SGM clients with criminal justice histories benefit from guidance identifying which people and spaces will be most safe for their SGM identities, histories of criminal justice involvement and homelessness, and other needs (e.g., sober living). Indeed, providers described how community connections were key to their success. Clients also benefit when providers can support them and step in quickly when they encounter setbacks, such as being denied an apartment, rejected from a job, or turned away from a social service agency.

It Is Important to Understand How SGM Identity, Homelessness, and Criminal Justice Involvement Are Intertwined in Complex Ways

Supportive housing clients with SGM identities and criminal justice histories have complex needs, given that all of those experiences and identities can influence each other. These dynamics are captured in the outer setting themes of cycles of marginalization and criminal justice system impact on SGM identities, but relate to the other CFIR domains as well. Staff need to relate to clients’ experience across all of the above dimensions (individual domain) and provide services that address different aspects of clients’ needs (intervention domain).

Discussion

Our interviews revealed a number of important needs for SGM people in supportive housing services with criminal justice histories. Themes related to important characteristics of the supportive housing interventions, individual staff and clients, organizations providing supportive housing services, extra-organizational contexts, and process of supportive housing implementation (i.e., the multilevel domains affecting program implementation; Damschroder et al., 2009). The themes represented a relatively even mix of implementation barriers and facilitators of supportive housing programs, highlighting the challenges experienced by SGM people with criminal justice and homelessness histories, but also the many strengths of these individuals and the providers supporting them. Ultimately, we identified five cross-cutting themes that spanned the multilevel domains, summarizing key considerations for supportive housing services and related policy that can help to maximize the identified facilitators while mitigating barriers.

Our use of qualitative methods, combined with a snowball sampling approach within Los Angeles County, limits our ability to generalize the details of our themes to other contexts. However, there are two reasons to expect that our findings have meaningful implications outside of Los Angeles County. First, Los Angeles County (and its constituent cities) have numerous resources in place to support SGM populations, including within the supportive housing services and criminal justice systems. For example, the Los Angeles LGBT Center has nearly 800 employees (Los Angeles LGBT Center, n.d.), offers supportive housing programs, and recently established the first comprehensive service center for transgender and nonbinary people in the United States (Trans Wellness Center, n.d.). Additionally, the Los Angeles County Men’s Central Jail operates a unit for men who have sex with men and transgender women (i.e., assigned male at birth; Dolovich, 2012). It seems reasonable to assume that other jurisdictions with fewer resources for SGM populations encounter similar, perhaps even more severe, challenges than our participants described—and would also benefit from adopting the approaches used by our participants to navigate discrimination and create affirming environments.

Second, our findings are largely consistent with research examining the factors that have led to overrepresentation of SGM adults among people experiencing homelessness (Bruce et al., 2014; Choi et al., 2015; Ecker et al., 2019; Wilson, Choi et al., 2020) and/or criminal justice involvement (Grant et al., 2011; Hanssens et al., 2014; Meyer et al., 2017; Vera Institute of Justice, 2020). More specifically, our findings expand on and provide more in-depth understanding of dynamics that have been documented in many other contexts, such as the complex interconnections among SGM identity and criminal justice system involvement as pathways to (and back into) homelessness (Tejani et al., 2014; Tsai & Rosenheck, 2012); the need for specific supportive housing intervention components addressing SGM identity and criminal justice history (Gillespie et al., 2017; Nyamathi et al., 2016; Reback et al., 2012; Salem et al., 2015); and the particular marginalization of transgender and nonbinary people (Grant et al., 2011; Vera Institute of Justice, 2020). Given these considerations, we now outline additional recommendations for policy and practice (i.e., supportive housing intervention programming).

Policymakers and Supportive Housing Organizations Need to Invest Resources in Developing Systems of Care That Address the Needs of SGM People Who Have Experienced Homelessness and Criminal Justice Involvement

This study did not provide a quantitative analysis of the system needs and capacity for these services in Los Angeles County, but our findings make it clear that the need is far higher than the capacity. In presenting our findings to county supportive housing provider organizations, we estimated that there is current capacity to provide affirming supportive housing for 1/6 to 1/3 of SGM clients (Dopp et al., 2021), based on providers’ estimates of program capacity and county estimates of SGM people in need of supportive housing. However, these estimates have not been empirically validated, and a more rigorous assessment of capacity is needed. Part of the issue, of course, is that there is not enough supportive housing stock (or housing of any kind) in the county or in much of California (Smith, 2004), let alone supportive housing that appropriately meets the needs of this vulnerable population. The solutions to that housing crisis are beyond the scope of this study, but certainly, there can be parallel efforts to maximize the proportion of supportive housing providers and related social services that are affirming for SGM people with criminal justice histories.

Policy solutions might include making staff training and support available for organizational development; dedicating resources and funds to organizations that demonstrate meaningful efforts toward affirming services; and developing and disseminating model policies that can help organizations understand how to structure their leadership and services effectively. Another solution could be partnerships between service organizations and service-oriented degree programs (e.g., in social work) to expand the workforce with supervised trainees while simultaneously increasing students’ preparation to work with high-need populations after training. In all efforts, it will be important to keep in mind the key components of supportive housing programs that were highlighted by our participants, such as comprehensive services and flexible timelines. Organizational leaders need to keep in mind that all individuals involved in a supportive housing program (staff, clients, leadership) make up its community, and thus contribute to creating an affirming or nonaffirming environment through how they engage with others. Selection and training of staff will contribute to the program environment as well as other important inner setting processes, such as team communication and community engagement.

Policymakers and Organizations Need to Take Particular Care to Ensure They Are Centering the Most Vulnerable and Marginalized Groups Within Improvement Efforts

This recommendation applies to transgender and nonbinary people, who were identified as being particularly marginalized in our interviews; it likely also applies to other SGM groups that were not mentioned by our interviewees, such as intersex and asexual people. There is a need for more organizations and initiatives that are led by and done for such marginalized groups. Moreover, improvement efforts must attend to the compounded experiences of racism endured by SGM people of color (Bruce et al., 2014; Choi et al., 2015; Wilson, Gomez et al., 2020). There are two reasons for centering the most vulnerable groups in efforts to improve supportive housing services. The first is an issue of justice, in that all people—regardless of social status—should have input into the solutions that are purported to benefit them. Too frequently, marginalized groups who lack representation within broader social justice movements are left behind. The second reason is an issue of effectiveness: marginalized groups have unique wisdom and strengths that they can contribute to broader efforts (Vaughan & Rodriguez, 2014). It is not simply that marginalized groups need the most help, but also that they can offer the most useful solutions. For example, the informal networks of SGM-friendly contacts described by the providers in this study offer a striking illustration of how marginalized groups empower themselves through their own communities and social structures. In terms of immediate mitigation, it was instructive that most incidents of discrimination described by our interviewees did not involve obvious, explicit mistreatment because of SGM identity or criminal justice history; rather, providers and clients were able to discern discrimination across patterns of behavior. Providers, with adequate support and training, can help clients understand what they are experiencing and navigate their responses to discrimination, whether overt or covert.

Formal Policy Barriers to Transgender People’s Employment, Housing, and Services Need to Be Eliminated at Government and Organizational Levels

Organizations that require identification should have policies addressing the fact that people may use a different name than what is listed on government-issued identification, for a broad range of reasons including (but not limited to) transgender identity. Service users should be given more flexible options for providing identification that do not require excessive self-disclosure or scrutiny of transgender people. Moreover, government agencies need to mitigate the complexity and cost of the process for obtaining a legal name change and updating the gender listed on official documents (e.g., birth certificate), and offer additional options besides male and female for indicating gender. Finally, the harmful practice of segregating congregate living settings (such as shelters and interim supportive housing) by the individual’s sex as assigned at birth needs to end. Such segregation is psychologically and, at times, physically unsafe for transgender people (Mottet & Ohle, 2006), which ultimately leads to transgender people declining shelter and contributes to high proportions of unsheltered transgender individuals experiencing homelessness. Overall, antitrans discrimination is codified within policies at a variety of government and organizational levels, all of which require modification.

Governments Need to Eliminate Policies and Practices That Effectively Criminalize the Consequences of Societal Marginalization and Discrimination in the SGM Population

Policies that impose criminal penalties on people living in poverty are a root cause of the cycles of marginalization described by our interviewees. Los Angeles County recently took a step toward decreasing its reliance on incarceration by developing a plan to close the county Men’s Central Jail (Ghaly, 2021). Recent analyses had found that more than half of the jail’s mental health population were appropriate candidates for diversion into community-based clinical programs rather than incarceration (Brooks Holliday, Pace et al., 2020) and that Black inmates were overrepresented among diversion candidates who remained incarcerated (Appel, Stephens, Shadravan, Key, & Ochoa, 2020). A similar analysis is not available for SGM populations in the county jail, but recent reports have highlighted the importance of prioritizing vulnerable SGM people for diversion whenever possible (Ghaly, 2021; Vera Institute of Justice, 2020)—especially since closing Men’s Central Jail will also mean closing its SGM unit. Therefore, it will be critical for governments to increase the resources and efforts allocated to nonpunitive alternatives to incarceration, such as supportive housing programs (Dopp, Brooks Holliday, & Hunter, 2020; Los Angeles County Alternatives to Incarceration Work Group, 2020; Vera Institute of Justice, 2020). Furthermore, reducing incarceration is only one part of the de-criminalization of SGM people’s lives—as another example, policing remains the largest expenditure in most large U.S. cities (Vera Institute of Justice, n.d.), and police harassment of SGM people is well documented (Mallory, Hasenbush, & Sears, 2015).

Increase Enforcement of Legal Protections Against Discrimination for SGM Populations

The U.S. Department of Housing and Urban Development (HUD) (HUD, 2021a) recently announced that it will begin enforcing the Fair Housing Act to prohibit discrimination on the basis of sexual orientation or gender identity when seeking to buy or rent a property (as well as related services, such as mortgages and housing vouchers). HUD also withdrew a proposed rule that would have allowed discrimination in homeless shelters on the basis on sexual orientation or gender identity (HUD, 2021b). These rulings affirm a federal commitment to reducing SGM discrimination in housing, but it will be important to allocate sufficient personnel, training, and resources to enable their enforcement at the local level. The fact that only about half of U.S. states have laws prohibiting anti-SGM discrimination in housing or employment reveals a continued lack of political will to protect SGM people in many jurisdictions (Warbelow, Avant, & Kutney, 2020). Moreover, policy efforts to reduce discrimination based on criminal history (also relevant to our focus population in this study) have revealed the challenges and pitfalls of many enforcement mechanisms—for example, barring employers from asking job applicants about criminal history increased job rejection rates among all Black applicants, regardless of history (Doleac & Hansen, 2020). Considerable care and effort will be needed to develop effective policies for promoting housing and employment among SGM people with criminal justice histories.

Conclusions

In addition to our policy recommendations, we recognize the need for additional research on these topics—in partnership with SGM communities (including those with lived experience of homelessness and criminal justice involvement) and policymakers. First, we encourage researchers to evaluate the impact of policy and program solutions on housing, employment, and criminal justice outcomes. For example, evaluations of SGM-specific interventions for people experiencing homelessness have had mixed success (Nyamathi et al., 2016; Reback et al., 2012) but offer important data about effective options to be tested, especially when considered alongside the findings of our study and data on supportive housing effectiveness more generally (Aubry et al., 2020; National Academies of Sciences, Engineering, & Medicine, 2018; Peng et al., 2020). Second, researchers can help develop more precise and rigorous estimates of system capacities and needs for SGM-focused supportive housing in a given community to inform future service planning and budgets. Third, continued research can help us to understand both the strengths and challenges of SGM people experiencing homelessness (i.e., avoid a deficit orientation; Vaughan & Rodriguez, 2014) and use intersectional analyses (Crenshaw, 2017) that account for compounding effects of multiple forms of marginalization within and in addition to SGM identities (e.g., racial identity, immigration status). Fourth, although the current study focused on SGM adults, it will be important to continue exploring the links between homelessness and juvenile justice involvement among SGM youth as well (Baams et al., 2019; Choi et al., 2015)—especially given that early intervention could help promote healthy development and prevent long-term problems.

Our findings and recommendations should be interpreted in light of several key methodological limitations. First, we did not conduct interviews with clients whose incarceration experiences involved immigration detention, as we did not have the resources to explore differences between those experiences and jail/prison. Second, we did not succeed in recruiting clients with certain SGM identities for interviews—notably intersex, asexual, Trans men, nonbinary, or lesbian women—and those same identities were rarely or never mentioned by providers. Our findings may not generalize to those groups, and certainly do not account for their unique needs—as demonstrated in rare examples of in-depth exploration, such as studies of Trans men’s pathways to incarceration (Rogers & Rogers, 2021) or experiences of sexual violence and pregnancy by men among lesbian women (Hodson, Meads, & Bewley, 2017). Third, our small sample made it difficult to identify needs of the SGM identities that were represented, beyond broad distinctions between the experiences of transgender and nonbinary clients versus cisgender SGM clients. Relatedly, we were also unable to differentiate needs by type of supportive housing (e.g., interim vs. permanent vs. Recovery Bridge) or by criminal justice history (e.g., prison vs. jail; length of incarceration). We likely did not reach saturation for these types of nuanced themes related to specific subpopulations, and more in-depth qualitative work is needed to explore such nuances. Fourth, our reliance on providers (who were primarily English-speaking) for client interview recruitment limited our ability to conduct Spanish-language client interviews, so we may have missed identifying important themes related to clients’ spoken language. Fifth, we might have obtained different results if we had not conducted this research during the COVID-19 pandemic—either because we could have conducted in-person interviews, or because interviewees might have had different experiences with supportive housing services. It did not appear that the pandemic had a major effect on the themes identified, but we did not specifically probe interviewees’ responses to determine how their experiences were different prior to versus during the pandemic.

In summary, this qualitative study identified important domain-specific and cross-cutting themes regarding the experiences of SGM people who have received supportive housing services in Los Angeles County and also have a history of criminal justice involvement. There are numerous important steps that policymakers and program administrators can now take to improve the experiences of and outcomes for this highly marginalized and vulnerable group. Through continued efforts that prioritize equity within our responses to homelessness, it can be possible to achieve a society in which all SGM people live healthy, happy lives and can fully share their strengths.

Acknowledgments

First and foremost, we thank the service providers who were generous with their time to participate in our interviews and assist with recruitment of clients. We also thank Susan Forrest of Los Angeles County Department of Health Services, Risa Flynn of the Los Angeles LGBT Center, and Cameron Varney of the Trans Wellness Center for their assistance with identifying provider organizations, and Mariana Marroquin of the Trans Wellness Center for her input into developing our interview guides and interpreting our findings. We appreciate Claudia Rodriguez of the Pardee RAND Graduate School for taking notes during the Spanish-language interview.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Rose Hills Foundation. The RAND Social and Economic Well-Being Division provided funds to support open-access publication of this article. ARD was an investigator with the Implementation Research Institute (IRI), at the George Warren Brown School of Social Work, Washington University in St. Louis, supported through an award from the National Institutes of Mental Health [5R25MH08091607] and the Department of Veterans Affairs, Health Services Research & Development Service, Quality Enhancement Research Initiative (QUERI).

Notes on contributors

Alex R. Dopp

Alex R. Dopp is a researcher at the RAND Corporation, an implementation scientist, and a clinical psychologist. He studies the use of research evidence, and related policy implications, to improve health and social services focused on vulnerable and marginalized populations. His research involves frequent collaboration with a variety of health policy, services, economics, and outcomes researchers, as well as service recipients and community stakeholders, within an interdisciplinary team science approach.

Sean McKenna

Sean McKenna is a research assistant at the RAND Corporation. His primary area of interest is urban economics, with an emphasis on implications for housing markets, homelessness, and land use. Furthermore, he has contributed to labor supply and personnel policy research.

Marylou Gilbert

Marylou Gilbert is a senior policy analyst at the RAND Corporation, Administrative Coordinator of the RAND Center for Housing and Homelessness in Los Angeles, and Project Manager with the RAND Pardee Graduate School, Community-Partnered Policy and Action Stream. She has an extensive background in community-based and community-partnered collaboration efforts and field research focused on immigrant, marginalized, and vulnerable populations. She is a bilingual (English and Spanish) trained social worker.

Sarah B. Hunter

Sarah B. Hunter is a senior behavioral scientist, Director of the RAND Center on Housing and Homelessness in Los Angeles, and professor at the Pardee RAND Graduate School. Her primary areas of interest are improving services for vulnerable populations and building community capacity for evidence-based program delivery. She has studied a number of supportive housing initiatives in and around Los Angeles County.