“The apartment is for you, it’s not for anyone else”: Managing social recovery and risk on the frontlines of single-adult supportive housing.

Abstract

This multi-method qualitative study examines frontline provider perspectives on consumer social relationships and barriers to social recovery in supportive housing programs for adults with serious mental illness. Thematic analyses show that guest and occupancy policies that enforce the “single” nature of single-adult supportive housing challenge consumer rights to self-determination in the realm of social recovery. Findings also highlight the ways in which providers act to reinforce and subvert these policies while mitigating risk in this service setting. Recommendations for enhancing the recovery orientation of supportive housing and implications for the design of the homeless service system are discussed.

Introduction

Social recovery: A superordinate dimension of mental health recovery

Mental health recovery has been broadly conceptualized as “a journey of healing and transformation enabling a person with a mental health problem to live a meaningful life in a community of his or her choice while striving to achieve his or her full potential” (U.S. Department of Health and Human Services, 2004, p. 1). One of the superordinate dimensions of mental health recovery is social recovery or “enhanced and meaningful relationships and integration with family, friends, and the wider community” (Whitley & Drake, 2010, p. 1248). Relationships with others impact individual wellbeing and shape one’s identity (Mezzina et al., 2006; Topor, et al., 2006), they can provide hope and encouragement critical to the recovery process (Spaniol, Wewiorski, Gagne, & Anthony, 2002), and while not all relationships are beneficial to recovery (Yanos, Rosenfield, & Horwitz, 2001), on the whole, connection with others is understood to be a vital component of mental health.

Accordingly, community and social integration of individuals with serious mental illness is now a major goal of mental health policy and the larger mental health system has been tasked with supporting social recovery for individuals in care (U.S. Department of Health and Human Services, 2003). Beyond providing support for social recovery, recovery-oriented services are also expected to be individualized, strengths-based and respectful of consumers’ rights to self-determination, choice, and risk-taking (Davidson, Rowe, Tondora, O’Connell, & Lawless, 2008). While much has been done to advance services in regard to clinical recovery and the management of psychiatric symptoms, less has been done to ensure that services for persons with serious mental illness also promote social recovery, or at the very least, do not impede social recovery.

Social recovery in homeless services

One sector of the larger mental health system in which barriers to social recovery have yet to be clearly articulated is in housing programs for individuals with mental illness. While the concept of mental health recovery and a recovery-oriented approach to service delivery has been widely promoted within the context of mental health primary services, there is little evidence that the broader recovery approach has been disseminated throughout the homeless service system where many individuals with mental illness receive support and rehabilitative services (Gillis, Dickerson & Hanson, 2010). Principles of recovery (e.g., consumer self-determination and choice) are considered important elements in regard to homeless policy, planning, research and services (Barrow, McMullin, Tripp, & Tsemberis, 2007; Tsemberis, 2010) and there is increasing recognition of the need to integrate this approach into homeless services (Gillis, Dickerson & Hanson, 2010; Sheedy & Whitter, 2009; Substance Abuse and Mental Health Services Administration, 2003). While individual programs and providers may discretionarily employ specific service models that support recovery, such as Housing First (Tsemberis, 1999), Wellness and Recovery Action Plans (WRAP) (Copeland, 2002), and motivational interviewing (Baer, Garrett, Beadnell, Wells & Peterson, 2007), system-wide integration of the recovery approach into homeless services has yet to occur. Studies have also shown that system-level factors, such as the structure of the housing continuum of care (Tiderington, 2017) and the funding of homeless services (Tiderington, 2015) can hinder frontline providers’ ability to deliver recovery-oriented homeless services on the ground.

Homeless service models have also demonstrated limited effectiveness in addressing service recipients’ social recovery. Supportive housing, a service model that combines subsidized housing and rehabilitative support for homeless individuals with special needs, has been proven to effectively reduce homelessness, increase residential stability, and reduce psychiatric hospitalization (Rog et al., 2014), but there is little evidence that this model enhances residents’ social integration. In fact, studies of social integration among recipients of supportive housing have generally shown that individuals make limited progress in social recovery, even after receiving permanent housing and support services (Stefancic, 2014; Tsai, Mares, & Rosenheck, 2012). Further, there is evidence that the move to an independent apartment may contribute to social isolation for individuals leaving a life on the street (Yanos, Barrow, & Tsemberis, 2004) and one of the most common complaints that those with serious mental illness have about living independently is that they feel lonely, isolated and without social support (Friedrich, Hollingsworth, Hradek, Friedrich, & Culp, 1999; Nelson, Hall, Squire, & Walsh-Bower, 1992; Siegel et al., 2006).

Homelessness itself has been framed as “a metaphor for disconnection from family, friends, and community” (Rosenheck, Bassuk, & Salomon, 1999) and studies of the social lives of homeless persons bear this out. Whether the cause or effect of homelessness, the depletion of homeless persons’ family and social networks has been well documented by researchers (Hawkins & Abrams, 2007; Padgett, Henwood, Abrams, & Drake, 2008; Padgett, Tiderington, Smith, Derejko, & Henwood, 2016). Not all formerly homeless individuals say that they want or are ready for relationships with others (Padgett, Henwood, Abrams, & Drake, 2008). Many report feeling generally ambivalent about social connectedness and actively distance themselves from others in order to avoid threats to personal resources and recovery (Stefancic, 2014). Given the compounded effects of serious mental illness and the trauma of homelessness, social recovery may then be a particularly daunting task for this population.

In many ways, services can act as a mediating force in an individual’s social recovery (Mezzina, et al., 2006), either by creating opportunities for social participation or by setting up barriers to the attainment and maintenance of connections with family, friends, and partners. Service providers are those individuals tasked with translating the federal mandate for recovery-oriented services on the ground and those in the frontline position to assist service users in their recovery of social relationships. However, little is known about if and how service providers actively address social recovery in housing programs serving formerly homeless adults with serious mental illness. Furthermore, the individual, organizational and systemic barriers to social recovery-oriented practice in this service setting have yet to be fully articulated.

This study examines how frontline providers of supportive housing for formerly homeless adults with serious mental illness view social relationships and the barriers to social recovery and social recovery-oriented practice in this service setting. Specifically, this study asks the following research questions:

• 1)How do frontline providers understand social relationships in the context of supportive housing?
• 2)What challenges, if any, do frontline providers experience to the delivery of supportive housing services that support social recovery?
• 3)What discretionary strategies, if any, do frontline providers employ in response to these challenges?

Method

Qualitative methods were chosen for the purposes of this study because these methods provide a way to examine research questions in-situ. To this end, this study incorporates not only what providers say (using in-depth semi-structured interviews), but also what they do in practice (using observations at the agency and ride-alongs with providers). Additionally, prolonged engagement in the field, with data collected over a period of almost three years (March 2011- February 2014), was used to build trust with participants, encourage candor and add rigor to this qualitative study (Padgett, 2016).

Data were collected as part of a larger National Institute of Mental Health (NIMH)-funded study on recovery in formerly homeless adults with co-occurring mental illness and substance abuse diagnoses who receive supportive housing services. A purposive sample of frontline supportive housing case management providers (N=35) was recruited through consumer participants from this study. Providers were selected from three supportive housing programs within two social service agencies in the same large urban area. These agencies were chosen because their housing programs conform closely to what Locke and colleagues (2007) describe as the two predominant models of housing services for homeless adults beyond emergency shelters – transitional and permanent supportive housing. One of the agencies in this study operated both a permanent and a transitional program; the other offered permanent housing only. The transitional housing program expected consumers to work on independent living skills and move on to permanent housing within two years of entering the program. The permanent housing programs provided housing to consumers without the expectation that they would move on from the program. All three of the programs employed a “scatter-site” approach, meaning that consumers’ apartments were scattered throughout the community in different locations, rather than concentrated in a single congregate setting.

Providers were interviewed about the respective consumer participant at three points in time – when the consumer first entered the program, after the consumer had been in the program six months, and one year into the program. But a single provider interview could include discussions about several different consumers in the study, each of whom could be at different points in time in their stay at the housing program and because provider recruitment was driven by consumer participation and subject to staff turnover, the distribution of provider interviews was rather complex. Some of the provider participants had several consumer participants on their caseload. Additionally, changes in caseload or staffing were also a frequent occurrence. This resulted in ten of the 35 providers in the sample being interviewed once. The remaining 25 providers had multiple interviews concerning multiple consumers over time. Overall, 35 providers were interviewed for a total of 84 interviews. The content of the interviews included the provider’s assessment of the consumer participant’s recovery and their views of the consumer’s experiences in the housing program, as well as their views on the barriers to and facilitators of the delivery of recovery-oriented supportive housing.

Targeted ethnographic visits to the program sites and ride-alongs with providers were also used to explore these research questions. In total, 106 hours of observations in the field were collected, resulting in 164 pages of field notes. The majority of observations were pre-arranged with staff and scheduled at times that were convenient for staff so as not to disrupt routine practice. Observations were timed to capture the typicality of day-to-day agency operations (e.g. weekly staff meetings), as well as special events (e.g. consumer graduations from the program). Researchers also observed providers while they conducted visits in consumers’ homes, employing Kusenbach’s (2003) “go-along method”. While data collection was focused on provider views and practice, at times observations also included consumers and others in the service setting. In these cases, no identifying information was recorded and private interactions were documented only if both parties provided verbal consent. Field notes were recorded immediately afterward whenever possible. Agency and funder policy documents were also obtained from participating agencies and online sources as needed in order to clarify participant-provided information. The University’s Institutional Review Board approved study protocols for this study and for the larger parent study.

For the purposes of this study, data were inductively analyzed in aggregate using Boyatzis’ (1998) thematic analysis. Interviews were transcribed verbatim and transcripts were entered into Atlas-ti software. Data were coded according to Saldaña's (2012) first- and second-cycle coding method. During the first-coding cycle, the author and another researcher from the parent study independently reviewed and open-coded these raw materials. We then undertook “consensus coding” together, a process used to establish agreement and increase the rigor and validity of coding in qualitative research (Padgett, 2008; Sandelowski & Barroso, 2002). Coded materials were then sorted according to sub-categories and aggregated into higher order themes.

Findings

Thematic analysis revealed several frontline challenges to consumer social recovery and the provision of social recovery-oriented practice found across all three supportive housing programs despite variation in program rules and expectations between programs. These challenges, along with discretionary strategies providers employed in response to these challenges, are detailed below.

Frontline Challenges

Enforcing the single unit.

One of the major challenges to social recovery and consumer choice described by providers and consumers in this service setting was the “single” nature of single-adult supportive housing outlined by agency and funder policies. While guest and visitor policies varied between the agencies, all of the programs in this study included single-use language in their rules of occupancy. This meant that the use and occupancy of the subsidized unit was intended solely for the recipient of services, or as one provider explained to a consumer, “the apartment is for you, it’s not for anybody else.” [2101]

Two of the three housing programs placed service recipients into private one-bedroom apartments using funding that subsidized the rent for one person only. The third program placed two to three service recipients together in a shared apartment, but with the understanding that the service recipients would be the only individuals residing in the unit. Expectations for single-use were communicated to residents via a Use and Occupancy agreement or rules of residency, as this agency’s agreement clearly stated, “The Apartment may be used and occupied solely by the Client. The Apartment may not be used as a primary residence by any other individuals.” Consumers were expected to sign this document at intake and agree to the single-use terms whether or not they had a significant other or roommate of his or her own choosing whom they wanted with them in the apartment.

The local occupancy law or “roommate law” applicable to the city where these agencies were located states that “it is unlawful for a landlord to restrict visitors to a tenant’s apartment. And it is unlawful for a landlord to restrict occupancy of an apartment to the named tenant in the lease or to that tenant and their immediate family. When the lease names only one tenant, that tenant may share the apartment with immediate family, one additional occupant and the occupant's dependent children” (New York Real Property Law §§ 235-f). But these housing programs got around this local occupancy law by maintaining the lease in the agency’s name – as the program director from one of the agencies explained during an observation, “We discourage any use of ‘lease’ or ‘rent’ language. Instead, we have them pay ‘program fees”’.

Providers were then in the frontline position to ensure that those in receipt of the individual subsidy were not in violation of the agency occupancy rules, as this provider said,

It is my responsibility to make sure that she is not letting anyone live in that apartment. And that was one of the things that I explained to her. ‘I’m not gonna be the police and come and check you out every week and show up in the evenings, so you can have who you want in here. But I’m telling you that if we find out that you’ve got people staying here, that’s going against your Use and Occupancy [agreement]. We can have you evicted because of that. [2118]

Agency guest and visitor policies further supported this message of single use, limiting consumers’ ability to freely entertain friends, family members, and significant others in the apartment when and if they desired. Both agencies exercised some discretion in how they defined their tenancy and occupancy standards. Site 1 had a more restrictive guest and occupancy policy, detailed here by a provider,

They can’t have visitors after 10:00pm and…you shouldn’t be having visitors in the apartment during the day. But as long as your guests come and go with you, that’s fine. As long as they’re not being with you in your room, that’s also okay. They can’t go in the bedroom… So, yeah, that’s all about visitors - that they not be under 18, that they not stay the night, and that you don’t have them in the bedroom… It’s all in the residency agreement, you have a visitation policy and everything else. [2101]

Even the act of consummating an intimate relationship could jeopardize one’s tenancy and result in eviction from this housing program, as this provider said, “You’re not supposed to have guests in the bedroom and he’s been having sex with somebody in the bedroom, so he’s been violat- you know they’ve been warning him a bunch of times. So therefore he has been discharged now, because he’s not abiding by the program rules.” [2112]

The other agency in this study had a more liberal policy – consumers were allowed under-18 visitors and overnight guests for a brief period of time - but they did not allow significant others to stay over long-term or move-in permanently unless the person had a legal relationship to the tenant and the agency had formally vetted the additional occupant’s ability to pay rent and/or their eligibility for an additional subsidy. But some consumers saw these stringent policies as hindering the development of intimate and social lives. During an observation of a community meeting in which consumers and agency staff would meet to discuss issues in the program, a male tenant raised his hand and brought the topic to light, “Dealing with a woman…that visitor policy, it makes having a sexual life hard!” A number of other tenants around the room nodded and began to raise their hands, but the consumer was cut off by a staff member and told to “get the conversation back on track.”

Negotiating social lives and program rules.

Frontline providers described a juggling act between honoring consumer choice and ensuring adherence to program rules. The act of striking this balance then entailed “a level of negotiation” wherein consumers would need to find ways to fit their social lives within the rules of the program, as this provider explained,

I think as people age they don’t feel the need to be in a relationship. They feel very satisfied being single. That was a part of their life before and they’re really not thinking about it. But some people, it’s really, really, a big issue, you know? And being married, being in a relationship is very important. Family, children, all of that. And who am I to come in your life to tell you everything you believe is wrong? But if you’re going to stay here, you’re going to have to work with us. We’re going have to have some level of negotiation. [2114]

Several providers talked about how the desire for social and intimate relationships is human nature, pointing out that consumers “are people too”, but they also qualified these needs within the bounds of the program rules.

They’re humans, they have needs too. Everyone needs their better half, you know?

…I like them to understand that it’s okay to have a girlfriend. There’s no rule in the [program] that says you cannot have a girlfriend. You cannot have a guest overnight, but you can have a girlfriend. It is for your safety, because if they have overnight guests they might forget about taking their medications and all that. But if you have your girlfriend and she visits you every now and then, and you visit her every now and then, it makes you get happy, you know? [2134]

Providers then had to juggle accountability to funders’ restrictions and agency policy with consumers’ needs for privacy and social recovery. Providers reported strategizing with consumers around how to enact this negotiation, as this provider said, “A lot of people believe ‘home’ is having a lot of company and having a good time. And from the gate, that’s going to be a problem. So let’s have this conversation, because we have to develop a different, um, strategy, for you to have company and feel fulfilled without it becoming a problem.” [2114]

In recognizing that tenancy rules could hinder social recovery, some providers advocated to those with greater discretionary power for more flexibility around enforcing occupancy and visitor rules, negotiating on behalf of consumers and their social lives.

He has a girlfriend who he now wants to move into the apartment with him. She lives like in a three-quarter house and he wants her to become an official tenant in the apartment. That can be a little tricky because if she becomes a tenant and their relationship doesn’t work out, then we can’t get her out for him. So I went back in and reviewed [the Use and Occupancy agreement], because my supervisor said ‘No, that’s not going to work.’ Being the social worker that I am, I don’t really like that response. I think that we have to be fair and if it’s going to get complicated, well I’m sorry, people are complicated and we’re going to have to deal with the complications. [2131]

This act of negotiating social lives within the program rules was also observed on a ride-along with a provider as we went to see a consumer whom the provider described as “a single guy, who doesn’t like living alone, so he picks up ‘strays off the street’ because he’s lonely”. She indicated that this consumer had several individuals who he would let stay the night in his apartment, but as she said, this was in violation of the Use and Occupancy of the unit. When we met with the tenant, the provider reiterated to him that he can’t let people stay in his apartment, that he’s risking eviction, and that he has to stop letting people sleep over. He responded by saying, “People need people to live. People need people to not go insane.” Her suggestion to him was that if he was lonely, instead of inviting people in from outside to live with him in his apartment, he should engage in a program-related activity and attend the agency’s art group or photography class.

Determining a fine line.

While providers made efforts to stay flexible to consumers’ needs for social and intimate lives in supportive housing, they also described “a fine line” between positive socialization and dangerous socialization, between a healthy social life and an eviction risk, and this line was often difficult to discern.

He was never sociable…and I guess the fact that he might be using, he became very sociable. Once he started using, he had the local crackhead, he had the local prostitute…they started leaving their clothes in his apartment. They know our hours and he knows our hours, so he’ll tell them, ‘Don’t come at a certain time, because my counselor’s coming.’ But once they’re gone, you know, he had all of their properties and stuff in the apartment … so he became more sociable, which is good, but that’s not how we wanted him to do it. (laughs) [2132]

Many consumers were described as vulnerable to exploitation, which made the transactional nature of social relationships a potentially risky endeavor. “You know they have mental illnesses, so these people take advantage of them. They know we’re a program, they know these people get money every week, they know that they’re a little impaired, and so it becomes easy to take advantage of them and get them involved in it, so it becomes a fine line.” [2113]

Formerly homeless consumers contending with poverty and addiction at times “brought the street indoors.” Consumers let others into their homes in an effort to accommodate still-homeless friends, in order to offer housing in exchange for drugs, or to gain a source of income by subletting the unit to others. This complicated matters, making the line between positive and negative social relationships that much blurrier.

He had drug dealers in the apartment and I had no knowledge of that. I mean the last home visit I did, it looked so home-y. He was cooking a meal for his girlfriend and another guy who was there…it looked very domestic. But underneath it something else had been going on. Because that apartment was then later raided by the SWAT team after he was incarcerated. They arrested two drug dealers in the apartment, armed! …So [the agency] was really like ‘You trashed an [agency] apartment, we’re liable for this. How are we going to re-house you again and have you damage property and ruin our relationships further with other landlords?’ So, it was tricky. [2131]

Even family reunification became a potentially dangerous venture that could jeopardize a consumer’s housing, as this provider explained,

I’ve known situations where the family said, ‘Oh, we want to know where she live. We want to visit! Can we come over and we all can meet?’ ‘Sure. That sounds like a good idea.’ And then it turned out that the family found out where the client lived, were visiting, stopping by, staying over—and then eventually, they kicked the client out of the apartment and took over! [2114]

Dealing with dangerous others and sitting ducks.

The realities of housing individuals with active addiction or other struggles with recovery meant that providers needed to stay vigilant to consumers’ associates and significant others in order to protect the person and their apartment. Consumers, lonely or vulnerable, were described by providers as “easy prey” or “sitting ducks” for exploitation, as this provider said, “This person is just a sitting duck, and if we put this person in that apartment, we know what is going to happen…and you see it all the time…she would like to have friends. But as I said, she’s easy prey.” [2111] Likewise, consumers’ associates and significant others were often framed as “dangerous others” - those who could act to exploit consumers and their access to housing and potentially cause their eviction.

If these others were found in violation of occupancy rules or they engaged in illicit behavior themselves, whether or not the consumer was also a participant, the consumer’s tenancy in the program could be jeopardized. As this provider said, it would only take one friend or family member, one significant other or friend of a friend, to end it all.

He has family that visits his apartment. He has a group of friends that he keeps very distant from the team, which is fine. I think part of that though is when you come from a street homeless background, what tends to happen is you bring one friend, let’s say from a shelter, and they see how you’re living, and then before you know it, everybody is like, ‘Oh, that’s the new place to hang out.’ So there’s always this concern that, you know, it only takes one person to sort of weasel their way into the apartment and then not leave, and then it becomes really problematic. [2104]

Allowing even one person to enter a consumers’ home carried with it a degree of risk, as dangerous others were often the downfall of a consumer’s success. “A lot of them fail because boyfriends come in, girlfriends come in. A lot of them fail because… well it’s mostly because people decide to move in with them. Or they’ve invited people to move in.” [2114] Or as this provider said, “All of our consumers who uses drugs, substance users…it happens to them. We have a few who have had apartments, lost the apartments because of it. …And all because they met a few friends. And their friends brought their friends, and their friends brought their friends, got out of hand.” [2111].

As the frontline enforcer of guest and occupancy rules, some providers acted to minimize the risk of eviction by discouraging consumers from developing social relationships. “I haven’t seen him meeting friends, but he’s friendly. He talks to people in the street. I just tell him to be careful with that. Like not to let strangers come to his apartment.” [2112] And some providers enacted closer monitoring of consumers when others came in, as this provider explained, “You got friends coming in 2:00, 3:00 in the morning, we don’t know who they are, they sleeping in the rooms. We’re going to check you, see if you been using, because we don’t know your friends. For things like that, I’ll give you an opportunity, but once I see that… then I have to step in and be a little stricter, a little stronger with you.” [2117]

When to step in and when to stand back.

In the end, it was frontline providers who were in the position to monitor for violations of tenancy, to make the call around when to step in and when to stand back. They had to determine whether others were helpful or harmful, whether the person stayed too long or too often, weighing a consumers’ right to risk and social recovery with the possibility that the consumer could lose their housing or worse.

In making this call, providers described two points at which the line had been crossed and it was time to step in and intervene: 1) when the person was at risk; or 2) when the unit was at risk. In the first case, providers said they would act in response to the person being kicked out of their own apartment, “I only enforce [the Use and Occupancy agreement] when it comes to the point like, there was a very low-functioning guy and a girlfriend that like moved in and kicked him out of the apartment. Then you have to enforce. …If their housing becomes in jeopardy, then we have to do something about it.” [2139]. And they acted in response to the person putting themselves or others at imminent risk, as one provider detailed here,

You have your friend in the apartment, and she’s high, and you decide to go out and she decides to cook and she falls asleep. I’m walking into the building with another counselor and there’s smoke coming out of the apartment and she’s sleeping…open up the door, she’s in the bed half-naked, cooking noodles and burning down the apartment and he’s outside! Something that dramatic, yeah, you’d be administratively discharged because you’re putting a stranger, the program, the building, other people lives in jeopardy. She could have killed everyone in the building. And that’s dangerous… About two weeks later he got administratively discharged. [2125]

Providers also acted to protect the agency’s housing stock and relationships with landlords, as this provider said, “You know if the cops get called into the apartment or if the building manager files complaints we can lose that apartment, and…we can’t afford to be losing apartments” [2113]. Or as this provider said, “We have about 20 apartments in that complex. If it wasn’t taken care of, then we would have lost all these apartments. Where are you going to put these guys? You’re putting people in jeopardy of losing their housing! [2125]. The loss of units was a concern at both agencies and even in the more lenient of the two agencies, there was a certain point at which consumers would be moved and re-housed in a different building if violations occurred. “I think it’s like three times and then—Like, you know, if we talk to you once, it’s not working. We do a contract with you, it’s not working. And then I consistently hear things from the landlord and from the police, we can’t do this anymore, because then we’re not going to be able to have a relationship with that landlord again.” [2131]

Discretionary Strategies

In the face of these frontline challenges, frontline providers enacted several discretionary strategies to determine when and if consumers were in violation of the tenancy agreement and they developed work-arounds to accommodate consumers’ social lives despite tenancy rules and restrictions.

Surveilling the unit and doing drop-ins.

On a ride-along with a provider, before we went to the consumer’s apartment, we went directly to the management office of the building where we watched surveillance videos of the foot traffic in and out of the person’s apartment. In order to monitor for tenancy violations, providers had to develop ways to track the severity of infractions in order to assess for when to step in and when to stand back by surveilling the unit. In this scenario, the provider said that she also used the act of surveilling the unit to appease building management. She explained to me as we left the building manager’s office that she sits through these meetings to “humor him,” giving him the impression that the agency would eventually do something to stop the individuals coming in and out of the consumer’s apartment and “buy the consumer time”.

Providers also tracked the severity of consumers’ occupancy or tenancy violations by “dropping in” to catch them in the act, as this provider said,

It’s hard to regulate [the visitor policy]. Most of the times, because we have keys to the apartment, on the weekends, at night, in the mornings, sometimes we do morning coverage…so we can come in almost any time, without notice. And that’s usually how we catch that going on. If you walk in and there’s someone there that shouldn’t be there, obviously, that’s not supposed to happen. [2133]

On another ride-along, as the provider and I approached the consumers’ apartment after buzzing his door, the consumer abruptly left, rushing past us down the stairs. The provider turned to me and said “He often tries to pull this when he has someone in the apartment”.

So the provider “keyed-in” to the apartment, letting himself in without the consumer’s permission, in order to catch the consumer’s girlfriend lying on the couch watching television.

Enacting a “man in the house” rule.

They also enacted, what one provider called a “man in the house” rule, referencing the practice of early welfare workers looking for a man in the house whose presence would indicate an additional breadwinner, ultimately putting the recipient's welfare benefits in jeopardy. Providers talked about monitoring for any sign of another individual whose presence as a live-in guest could result in a client’s eviction and the loss of their subsidy by explaining to consumers “I don’t want to see any males, I don’t want to see no Old Spice, no boots.” [2114] Or as this provider said,

He knows that she is not allowed to live in the apartment. She can come like every day if she wants to and hang out with him. That is not something we can control, but she is not allowed to sleep in the apartment. And when I go there I always try to look around to see if I see clothes, like female clothes or female products around the apartment. That’ll tell me like, okay, she is living here. [2115]

Items left might be the only indication that the person’s significant others were there in violation of the occupancy rules, the only way to catch someone in the act. “As long as they have them out before 9:00 or 8:00, when we start…it’s hard to catch it, but when we start seeing women’s clothing and, you know, things like that, it’s like, okay, come on! (laughs)” [2101]

Turning a blind eye.

Providers also strategized to facilitate consumer goals of having a social or intimate life by “turning a blind eye” to relationships they discretionarily deemed “healthy” or “beneficial” or in general “less of a risk” to consumers’ tenancy and wellbeing, as this provider said, “As long as I don’t see it, I’m not going to question it. As long as I don’t have indicators that let me know when I visit, you know, the man’s side versus the woman’s side (laughs) and different things, I’m not going pursue the subject.” [2114] And this was a discretionary call on their part, as this provider explained,

It always goes from client to client when it comes to that Use and Occupancy agreement. If it’s a beneficial relationship… I’ve seen so many where you almost feel like thank god they have somebody at the end of the day…not to do my job, but somebody that understands them. Is that always the case? Very far and few. But for the few that work out, I really don’t want to rattle their cage too much, you know? [2114]

Another provider said,

I have another lady who the kids sleepover. She’s got bunk beds! They’re all in foster care during the week. I have a lot of older clients whose kids are around, spend a couple nights there, support them, help them move. Help them with this, that. They’re not supposed to be - because this is single-occupancy. But I think there are a couple cases where the kids are around enough where you may call it ‘living with them’. But it’s not official. [2139]

Even if these individuals were beneficial to consumers, they were still in violation of the single-occupancy restriction, but both providers and consumers worked to minimize the chances of this violation coming to light, as this provider said,

The reason why I probably don’t really push the subject, I don’t drop in, is because I know that’s her support. [Her common-law husband] is her path to wellness, and he helps her a lot. He knows that he cannot show signs– that he can’t be in the bed relaxing while I’m there. He knows he has to toe the line. And I think that’s their way of showing respect to the rules. [2114]

Encouraging relationship (and tenancy) formalization.

Providers also encouraged a work-around to the official occupancy policy, by encouraging consumers to get married and formalize relationships with partners whom they wanted to move in, because this would then put the agency in the position to advocate on the consumer’s behalf for a second subsidy, as this provider said “We can just tell people ‘get married’ and you know, we’ll honor that. But then we have that dilemma, you know? Not everybody wants to get married, but they want be in that kind of a relationship. [2114]

Technically, another person could not take advantage of the subsidized apartment unless that person was “on the books” with the agency and legally bound to the service recipient, as this provider explained,

We tell the clients ‘You can have guests. They can stay for a week! They can probably stay for two weeks! But if they are living with you, they must also sign a sublease and they must also pay rent’ …depending on their relationship. Like if it’s just a friend of his - no. But let’s say [the consumer] got married. Then yeah, [the wife] would come in, she would sign a sublease, she would send in her pay-stubs or her award letter [from Social Security] and she would have to pay rent. [2121]

This led providers to encourage the formalization of partner tenancy and of intimate relationships, even in cases when the consumer preferred not to do so, as this provider said,

Use and Occupancy agreements, you cannot live with somebody unless you’re legally married. …That was kind of problematic for [this consumer], because she is in a relationship that she regards as lifelong, consistent, but she’s not willing to get married. So therein lies the problem. She wants the privilege of—because we have clients that live together. They’re married, they have documents to prove that they’re married, and so they live together. Each one of them separately pays a portion of the rent, based on what they can afford. But, in their case, she’s not willing to get married, so that was problematic. [2114]

Discussion

Frontline providers understood consumer social relationships to be a “double-edged sword” with potentially favorable or dangerously unfavorable consequences. They viewed significant others as a possible risk to the agency and the individual, but they also saw these relationships as potential supports for recovery. In response to this ambiguous line between healthy and unhealthy relationships, these “street-level bureaucrats” enacted their own discretionary strategies (Lipsky, 2010) to determine when and if a consumer’s social relationship had become too great of a risk and when and if to intervene.

At the individual-level, findings from this study highlight the need for additional clinical supports and supervision for those grappling with challenges to social recovery on the ground and the risks involved in housing individuals in recovery. As the providers in this study note, an inherent tension exists between one of the fundamental tenets of recovery – allowing individuals with mental illness “the dignity of risk and the right to fail” (Deegan, 2006) – and the very real risks associated with letting “dangerous others” into program apartments. Additional training and supervision around implementing a harm reduction approach and the use of motivational interviewing (a client-centered approach in which providers work with clients to facilitate healthy decision-making and behavior change) could provide frontline workers with more of the clinical tools needed to maximize consumer self-determination while minimizing risk in these settings, ultimately enhancing the overall recovery orientation of these services.

Findings also suggest that organizational- and system-level barriers to social recovery exist within the homeless service system. Both providers and consumers described agency and funder-dictated guest and occupancy policies as barriers to social recovery. These policies were seen as undermining consumers’ rights to choose those with whom they want to live and to freely entertain friends, family and intimate partners in their own private living space. Single-use regulations promoted surveillance and oversight of consumers’ social lives, creating a dual-role for frontline providers in this service setting - as both clinical supports and enforcers of tenancy rules. Although research on supportive housing has shown that surveillance in supportive housing fosters the conditions under which residents can feel safe and exert control over their own lives, it has also shown that consumers view these surveillance strategies as intrusive (Parsell, 2015).

The presence of single use and occupancy policies in supportive housing programs also underscores how the current structure and funding of homeless services perpetuates a false dichotomy in services between homeless families and single-adults as two separate and distinct groups. Much of the homeless service system in the United States is structured in this way, operating and funding two separate networks of care: one for unaccompanied (i.e., single) adults and another for families (i.e., childless couples, pregnant unaccompanied women, or one or more adults with children). Upon intake, homeless persons are typically filtered through the system based on their current family status. Homeless women whose children reside with others may not qualify for family shelter placements or family supportive housing services unless they are pregnant or coupled at the time of intake. Likewise, homeless men who enter the system without a partner or children are typically tracked to services for “unaccompanied single adults” like the supportive housing programs in this study.

Yet, these two populations are not as discrete or static as these artificial boundaries might suggest (Barrow & Laborde, 2008) and this tracking system can lead to problems down the road if and when a single person wants to reunite with their family or move in a partner, roommate, or other family member. Surveys conducted throughout the era of modern homelessness have repeatedly shown that the vast majority of women residing in homeless services for single adults have children who do not reside with them (Crystal, 1984; D’Ercole & Struening, 1992; DiBlasio & Belcher, 1992; Smith & North, 1994; Zlotnick, Tam, & Bradley, 2007) and many parents, when homeless and transient, voluntarily place their children with family, friends or in foster care. Others lose their children to the foster care system just because they are homeless (Shinn & Weitzman, 1996). Still others become separated from loved ones upon entry to the homeless system, as the rules and regulations of shelters (e.g. the exclusion of men and adolescent boys from some family shelters) can inadvertently separate families and disrupt romantic relationships. It can then be assumed that individuals, upon receipt of stable housing, may seek to rebuild these relationships.

Individuals who enter supportive housing as unaccompanied, single adults are unlikely to receive a unit or subsidy large enough to accommodate additional occupants over time and limited resources challenge supportive housing agencies’ ability to locate affordable housing units and subsidies which can be easily altered to meet consumers’ changing needs in this area. Although the structure and funding of supportive housing services has begun to adapt with some single-adult programs offering assistance to clients who wish to move to a larger unit or accommodate another occupant, there remains a need for greater flexibility in services. Permanent supportive housing programs specifically targeted to families is a relatively new incarnation of the supportive housing model that begins to address some of these needs, but this approach may further perpetuate this false dichotomy by continuing to set these two groups apart as if they were static in nature. Unless larger “family” units are readily accessible to all, as needed, over time, and housing programs utilize a more flexible definition of “family” (including non-married partners and roommates of one’s own choosing) in their use and occupancy standards, the federal goal of a recovery-oriented system of care may remain unrealized.

Table 1.

Provider Participant Demographics

		Site 1 PERM/TRANS N (%)	Site 2 PERM N (%)	Total N (%)
Gender		24	11	35
	Male	10 (42)	3 (27)	13 (37)
	Female	14 (58)	8 (73)	22 (63)
Race/Ethnicity
	White	1 (4)	4 (36)	5 (14)
	Black/African American	20 (84)	5 (46)	25 (71)
	Other	2 (8)	2 (18)	4 (11)
	N/A	1 (4)		1 (3)
Employment Tenure
	<1 year	8 (33)	6 (55)	14 (40)
	1-3 years	10 (42)	2 (18)	12 (34)
	>3 years	5 (21)	3 (27)	8 (23)
	N/A	1 (4)	0	1 (3)
Highest Degree Completed
	Graduate	1 (4)	7 (64)	8 (23)
	Bachelors	13 (54)	2 (18)	15 (43)
	Associates or less	8 (34)	1 (9)	12 (33)
	N/A	2 (8)	1 (9)
Prior Experience
	Yes	19 (79)	8 (73)	27 (77)
	No	4 (17)	3 (27)	7 (20)
	N/A	1 (4)	0	1 (3)