|Home | About | Journals | Submit | Contact Us | Français|
Housing is a crucial issue for adults with severe mental illness and co-occurring substance use disorders, as this population is particularly susceptible to housing instability and homelessness. We interviewed 40 adults with dual disorders, living in either supervised or independent housing arrangements, to examine housing preferences, decision making processes surrounding housing choices, and perceived barriers to housing. We found that many respondents indicated their housing preferences had changed over time, and some clients related housing preferences to recovery. Although the majority of clients preferred independent housing, many also described benefits of supervised housing. Clients’ current living situations appeared to be driven primarily by treatment provider recommendations and availability of housing. Common barriers to obtaining desired housing were lack of income and information. These findings have implications for supported housing models and approaches to providing housing for clients.
Research on which type of housing is most appropriate for adults with severe mental illness has been inconclusive (Fakhoury, Murray, Shepherd, & Stefan, 2002; Leff et al., 2009; Siegel et al., 2006). In the current study, we specifically focused on the client perspective to gain a deeper understanding of housing preferences- what clients value about housing and the reality of how housing choices are made. We focused on adults with a severe mental illness and a comorbid substance use disorder, i.e., a dual diagnosis, because they constitute a vulnerable group that is susceptible to housing instability and homelessness (Drake & Mueser, 2000).
Two predominant housing models are the residential continuum and supported housing. While the residential continuum gradually transitions individuals to community living through placements in progressively less restrictive and less intensively staffed housing arrangements (Ridgway & Zipple, 1990), supported housing is broadly defined as independent housing in the community with provision of mental health support services (Hogan & Carling, 1992; Rog, 2004; Wong, Filoromo, & Tennille, 2007). The debate surrounding these two housing approaches involves two issues. The first is balancing client preference with client need, which is not always synonymous. For example, some studies have found that clients who strongly prefer independent housing are more susceptible to subsequent housing loss (Goldfinger et al., 1999; Schutt & Goldfinger, 2000). The second issue is an underlying assumption that all clients prefer independent housing all the time (Allen, 2004; Carling, 1990). The majority of clients surveyed prefer independent housing; however, as many as 41% of clients are interested in other forms of housing as well (Tanzman, 1993). Independent housing is also a broad category that may include apartments and single room occupancies (SROs), which clients often do not view as equivalent (Tanzman, 1993). Furthermore, research has found that client preferences can change over time with different experiences (Schutt & Goldfinger, 2000). Many homeless dual diagnosis clients who are not interested in treatment cannot tolerate a highly structured, restrictive environment after living on the streets (Blankertz & White, 1990), and it can be expected that many of these clients would prefer independent housing. But as active efforts are made to recover, clients may then seek the supports and resources offered in supervised settings.
In addition to the client’s housing preference, other factors must be taken into consideration when examining housing options. One factor is financial; most clients are constrained by limited income (Mechanic, Bilder, & McAlpine, 2002; Yeich & Mowbray, 1994). Another factor is the influence of treatment providers and family members (Friedrich, Hollingsworth, Hradek, Friedrich, & Culp, 1999; Srebnik, Livingston, Gordon, & King, 1995), who often recommend housing with more structure than clients prefer (Goldfinger & Schutt, 1996; Holley, Hodges, & Jeffers, 1998; Piat et al., 2008). Other barriers to housing include criminal histories (Clark, 2007) and stigma surrounding mental illness (Robert Wood Johnson Foundation, 1990). Solutions such as integrating evidence-based practices have been proposed to improve housing services (Kloos, Zimmerman, Scrimenti, & Crusto, 2002; Tsai, Salyers, Rollins, McKasson, & Litmer, in press), but a better understanding of the factors and barriers to desired housing is warranted.
This study was part of a larger study of housing preferences that included surveys with 103 participants at Thresholds, a large psychosocial rehabilitation agency that serves over 5,000 people diagnosed with severe mental illness in the Greater Chicago area. The current study focused on 40 of those participants who were interviewed, 20 each from supervised and independent housing provided by Thresholds. To be eligible, clients had to have a history of homelessness and a dual diagnosis, defined as having a severe mental illness (schizophrenia-spectrum disorder, bipolar disorder, major depressive disorder) and a substance use disorder (any substance dependence or abuse diagnosis). The supervised housing sample consisted of clients from two residential programs described elsewhere (Davis et al., 2006; Mayes & Handley, 2005; McCoy et al., 2003). The independent housing consisted of clients living in various independent housing throughout the city, who were receiving comprehensive case management services. All study procedures were approved by institutional review boards at IUPUI and Thresholds.
Twenty clients in supervised housing were randomly selected (10 clients from each residential program) and all clients approached agreed to participate. All 160 clients in independent housing who participated in the questionnaires from the larger study were mailed a flyer inviting them to participate in the semi-structured interview. The first twenty clients who responded to the flyer were included. The participants from independent housing consisted of 14 clients living in apartments, 5 in SROs, and one residing with family; these subgroups were analyzed separately as they were viewed as different types of independent housing. All participants received $5 compensation for participation in this part of the study. The majority of participants were Black males with a mean age of 46.3 years (sd= 7.7) and a mean residential tenure of 33.2 months (sd= 31.4) in their current housing. The number of different prior residences in the past 5 years ranged from 1 to 12.
We used a semi-structured interview, guided by 7 broad topics: 1) What do you think of your current housing? What do you like? What don’t you like?; 2) How did you come to live here? How much choice did you have? If you had choice, what were your main reasons for choosing to live in your current housing?; 3) What type of housing would you prefer right now and why?; 4) If you are not in your preferred housing, why not?; 5) Do you think your housing preferences have changed through your recovery? How so and why?; 6) What do you think about the people that you live around? Do they help you in your recovery?; and 7) What type of housing do you see for yourself in the future and why? Generally, clients were encouraged to express thoughts and opinions freely with interview questions used merely to direct responses on certain topics. Interviews were conducted by the first author in each participant’s home. Each interview took approximately 45 minutes. Interviews were audiotaped, transcribed, and checked for accuracy.
Client interviews were analyzed using content analysis (Weber, 1990) informed by grounded theory (Glaser & Strauss, 1967). First, all of the transcripts were read and concepts were written in memos. Through an iterative process involving discussions, readings, and modifications, memos and ideas were organized into general categories and subcategories that were compiled into a codebook. Two coders (first author and fourth author) independently coded the content of 20 identical transcripts- 10 from independent housing and 10 from supervised housing- then discussed and reached consensus on any discrepancies. After 20 transcripts, it was determined that adequate inter-coder reliability was established and the codebook was sufficient. The remaining 20 transcripts were split and independently coded by the two raters. The final codebook consisted of 10 general coding categories and 82 sub-codes. Inter-coder reliability was calculated, before consensus, on the last 12 (30.0%) transcripts that were coded by both coders. There was adequate inter-coder reliability with an overall Kappa value of .77, and coding categories ranging from .65 to .83. Coding patterns were examined, memos compiled, and transcripts reread to identify general themes. The bulk of the analyses focused on the 14 clients in apartments and 20 clients in supervised housing, however specific findings on SROs were also noted. The computer software, Atlas.ti (Atlas.ti Scientific Software Development GmbH, 2008) was used in coding analysis and the Coding Analysis Toolkit (Shulman, 2007) calculated inter-coder reliability.
Certain housing characteristics were enjoyed by clients in both supervised and apartment housing. In particular, clients enjoyed having their own space and keys so there was a sense of ownership, their own bathroom, and a kitchen for cooking. Clients in supervised and apartment housing both liked the independence and autonomy they had. For instance, more than half of the clients in both types of housing equally reported the freedom of being able to ‘come and go as I want’ (independent client 3) and ‘do anything I want’ (supervised client 2). While there were similarities in perceived independence and autonomy, there were some differences in privacy. Nine of 14 (64.3%) clients living independently in apartment housing described privacy in their housing, while only five of 20 (25.0%) clients in supervised housing mentioned privacy and three (15.0%) complained about regular staff inspections.
The sense of community was also different for people living in apartments versus supervised housing. Most of the clients in apartments either interacted only with other clients who were also living there or they kept to themselves. In contrast, 18 (90%) clients in supervised housing reported a sense of community among the other tenants and receiving peer support.
‘You never get lonely. That’s the thing, you know. Like if I was in a hotel room, you know, it wouldn’t be as many people. But you know, I like to be around people, I’m a people person, so you know, it’s kind of nice having so many people in one area that you can really talk to. We got our own community neighborhood right up in here. We don’t need to go outside, we got community. Look at all these rooms up in here. You know, they got certain things like bikes in the basement, they go bike riding, they go camping. There are a lot of things positive that goes on in the building’ (supervised client 15).
Overall, clients in both independent and supervised housing reported receiving good staff support, except for a few complaints in supervised housing regarding staff at the housing facility who did not respect confidentiality and/or did not always enforce the rules.
When discussing housing dislikes, several clients living independently complained of the poor physical condition of the housing; 5 (35.7%) clients in apartment housing and 4 (80.0%) clients in SROs talked about dirtiness, bugs, poor repair, and cracks in walls, while clients in supervised housing did not voice these complaints. Clients in supervised housing, however, did have complaints about tenants, despite also reporting camaraderie and peer support. Eleven out of 20 (55.0%) clients in supervised housing complained about other clients who used alcohol and drugs in the building, stole things, or were disruptive in some way. The author observed there were some “Do Not Disturb” signs taped to doors in supervised housing to corroborate what clients were reporting.
Almost all clients reported they had a choice about moving into their current housing, i.e., they didn’t feel obligated or coerced. In fact, only 3 reported having no choice because of a court order or miscommunication between staff. Although few clients felt forced into housing, many (16, 40.0%) described having few to no housing options. Among clients who reported having options, most reported only having 2 or 3 options with one client reporting the most at ‘5 or 6 different places’. Housing options were provided by treatment staff or through their own searching, and sometimes these were superficial options (e.g., choosing dirty halfway house or own clean apartment).
Clients often said they did not think or ask about other options, because either they did not care at the time due to their desperation or they did not realize there were other options. For example, when one client in supervised housing was asked why he didn’t get his own apartment, he responded, ‘I didn’t know about anything like that. I was offered just the one avenue. I didn’t know about any other thing, and at the time, it didn’t really matter, to get off the streets was good. You don’t think about things like that, especially if you don’t know those options are available. But you’re not picky at that point in time’ (supervised client 12).
There were a range of different pathways in which clients in both supervised and independent housing arrived at their current housing. It did not appear clients went through a thorough decision-making process when selecting housing. In addition, there was no obvious adherence to any particular housing model. Sometimes clients moved simply to leave an undesirable living situation (e.g., homelessness, housing dissatisfaction), and other times they moved as part of their recovery.
‘I asked my case manager at Thresholds to move me out because I got tired of staying there using drugs and wasn’t getting no better. And I didn’t want to keep on owing people money every week… I had to get out of there quick as possible because I got tired of living like that. I had that money every week but I wasn’t taking care of my functions like keeping my clothes clean, and my vinyl all soaked in deodorant stuff’ (independent client 11).
Clients moved from place to place based on what was available and where treatment providers suggested they go. As mentioned earlier, clients didn’t feel forced into any particular housing, but often there were limited options. Not surprisingly, treatment providers played a large role in where clients were living and clients relied on them to find housing. Client’s family members were another big influence; either through encouraging clients or being directly involved in referring and accompanying them to view housing. Negative experiences and perceptions of housing also played an influence on where clients chose to live. Five out of 14 (35.7%) clients in apartment housing expressed negative perceptions or experiences with group living, and did not want to live in that type of housing. Some clients clumped all different types of group housing together (e.g., halfway houses, group homes) and viewed them as the same, not wanting to live in any of them.
Barriers to housing were mostly reported in the context of seeking independent housing. Financial factors emerged as a major barrier, mentioned by 24 (60.0%) clients. As one client succinctly put it, ‘living circumstances is demographics ziplocked in economics’ (independent client 8). Financial factors included low income, waiting for government benefits, bad credit, and whether apartments accepted housing vouchers (e.g., Section 8 which is a federal rental subsidy). Other barriers to independent housing included criminal history and stigma. Six clients reported that their criminal history created difficulties with finding and being approved for independent housing. Stigma was only discussed by three clients, but it is worth noting here.
‘If you have a landlord that knows that you might have an illness, I got a feeling that they look at people differently so that pushes them to go wherever they can be accepted. There’s a hotel down here don’t accept people from Thresholds. I mean, I guess, somebody went in there and did a couple things wrong… That’s got to quit because people should be able to have the opportunity to live where they want to live. If they can afford the rent, live where they want to live without having a label on their door, you know, or a cross or something, written in blood to watch out for this individual’ (independent client 20).
The only barrier described by clients seeking supervised housing was the waiting list for some of the units. It was mentioned by seven clients seeking supervised housing, and also reported as a barrier by two clients seeking independent housing. Clients were often on waiting lists for months. In the meantime, clients described being nomadic and were hard to re-contact when it was their turn on the waiting list.
Of the 40 clients interviewed, 24 (60.0%) currently wanted to live in some type of independent housing (20 apartment, 3 house/trailer, 1 SRO) and 16 (40.0%) wanted some type of supervised setting (14 residential program, 1 nursing home, 1 ‘mental hospital’). The majority of clients who wanted independent housing wanted to live with the general population, but a handful of clients wanted to live around others who had mental illness or said ‘doesn’t matter as long as I’m comfortable’ (independent client 6). Thirteen clients (65.0%) living in supervised housing and 8 (40.0%) living in apartment housing were living in the exact housing situation they currently wanted (i.e., wanted to stay in the same room); none of the 5 clients in SROs were residing in their preferred location. Reasons for preferring independent housing were privacy, autonomy, independence, and space.
‘That’s my ideal situation to actually be self-sustaining, a regular part of the community, you know. I want a part of the community. I just want a place to live that I have the keys to, I can open the door, I can shut the door, I can answer the bell, not answer the bell, little bit of my own. This is where I’m at, you can’t mess with me, this is it, it’s mine’ (independent client 8).
Reasons for preferring supervised housing were structure, a substance-free environment, and staff and peer support.
‘Because I need policing. You know, I need policing, I need somebody to help me stay on the right track. And without their policing, I don’t know, I’d probably still be out there trying to sleep in the cold, or somebody else’s house where they didn’t really care about me. They just wanted me for the money once a month and the rest of the month, they talk to me like I was stupid or like I’m crazy’ (supervised client 18).
The type of housing clients currently wanted was sometimes different from what they wanted in the future. Except for one client, all wanted some form of independent housing in the future. Twenty-three (59.0%) wanted an apartment and 16 (41.0%) wanted their own house, condo, or trailer. One-fourth of all clients wanted to live with a future romantic partner or with family.
Clients were asked whether and how their housing preferences had changed over time. Twenty six of 40 clients (65.0%) described having different preferences at different times in their lives, equally distributed between clients in supervised and independent housing. Of those that said their housing preferences had changed, 17 (65.4%) described preferences changes in relation to their level of recovery. The remaining 9 (34.6%) described nonspecific changes in preferences related to housing experiences, life changes, or unidentified reasons.
In describing preference changes related to recovery, many clients reported they wanted supervised housing at some point in their lives to aid in their recovery from mental illness and substance abuse. When they felt stable enough in their recovery, they wanted independent housing.
‘Well, what’s hard about like being not in a group home and having my own apartment, I guess for me, it would be hard because I might get to thinking that I could, that I’m free to do whatever I wanted to. And for myself, I can’t speak for nobody else, I would end up drinking and getting high and stuff… It would be best to get in a group home like being in a place where people surrounded with mental illness and substance abuse, in a program like that. I think it would be best if I try that out first, and then if that worked out and I was in a program like that, and then if I graduated from a program and I was doing good, graduated and everything, I think then I could work on something like my own place. I think it would work out better if I work on that problem first’ (independent 11).
Other clients who moved onto independent housing retrospectively described recovery-related preferences.
‘I took the steps at that time, you know, I was very manic before so I needed the housing whereas I had my kind of people around me as support in order to snap me back into what I needed to do for reality… So I needed that support, I needed those groups. I needed that time where we all got in the medication line and we took all our medication. So that took me from one step to another step to another step, but you have to be willing to want to take some responsibility and do what you need to do for yourself’ (independent client 4).
Also telling was that 16 (40.0%) clients who did not explicitly verbalize recovery-related preferences spoke of how supervised housing may have or did benefit them in their recovery. Although they didn’t prefer it, they realized in retrospect that it was what they needed at the time.
This study examined what is important to dual diagnosis clients regarding housing, what types of housing they prefer, and what factors led to their housing. As expected, the majority of clients preferred living in their own apartment or house. However, housing preferences were not static, and over half of the clients reported that their housing preferences had changed over time, often relating these changes to their recovery. As clients felt themselves progressing, many wanted different types of housing, ‘depending on where you’re at in your stages of recovery or wellness with your mental health would dictate how your thoughts are about caring where you want to stay’ (supervised client 12).
Interviews revealed that almost all clients wanted some form of independent housing in the future, and one fourth aspired to live with a future significant other or family member. This was in contrast to the relatively isolated lives described by clients currently living in independent housing, consistent with previous findings (Friedrich et al., 1999; Siegel et al., 2006). This has implications for supported housing programs- increasing socialization may need to be a focus for clients living independently. Although nearly all clients wanted independent housing in the future, many described needing supervised housing at some point in their recovery. Many clients talked about how supervised housing provided structure and support that was helpful in their recovery. Clients who reported they had never wanted supervised housing acknowledged and in some cases, extolled the benefits of supervised housing especially ‘for starting out’ (supervised client 5). This has implications for the residential continuum model, which moves clients from supervised to increasingly independent settings (Ridgway & Zipple, 1990).
Supported housing programs that exclude group living arrangements (Hogan & Carling, 1992; Rog, 2004) may ignore the possible benefits of supervised housing. Research has shown dual diagnosis clients can benefit from supervised housing (Brunette, Mueser, & Drake, 2004; Goldfinger et al., 1999). In the current study, over half of the clients in supervised housing clearly talked about how they preferred supervised housing right now and despite their stated future preference for independent living, many clients were very satisfied living in supervised housing. Our findings suggest that an array of housing options may be needed. As other researchers have pointed out (Friedrich et al., 1999), giving clients a choice means ensuring housing alternatives exist. As the supply of low-cost independent housing and government funding for social programs diminish, the value of other housing alternatives should be considered.
One surprising finding was that clients in both supervised and apartment housing reported enjoying the amount of independence, ownership, and space their living situation provided. These similarities were unexpected given the finding that clients associate supervised and apartment housing with different housing characteristics. It may be that there are fewer actual differences between the two types of housing than sometimes assumed. Also notable is that SROs, although often considered independent housing, were viewed by clients as less than adequate and inferior to apartment housing. There were a limited number of SRO cases in this study, but future studies that evaluate independent housing may find it useful to make a distinction between apartments and SROs.
The residential programs in this study can be characterized as “damp” housing and it is unclear whether the results would generalize to “dry” housing where strict rules on substance use are enforced. Some clients had negative experiences with supervised housing and envisioned all group housing as living with roommates and having strict rules- many believed they could only have their own room, kitchen, and bathroom in independent housing. As in this study, supervised housing units can vary greatly on these features, but clients often do not realize this. Corrigan et al. (2008) presents a typology of various residential arrangements that may be useful for providers presenting options to clients. However, the specific features of housing should also be carefully discussed with clients beyond presenting them categorically as the categories may not be so distinct.
Clients often did not have a rational decision-making process in selecting housing. Most times, decisions regarding housing were based on suggestions from treatment providers. Because we did not interview treatment providers, we do not know their perspective. But the influence of treatment providers may be important in light of previous findings that what treatment providers see as good for clients is often at variance with what clients want (Holley et al., 1998; Piat et al., 2008). Shared decision-making has been found to be crucial in increasing treatment adherence and endowing patients with responsibility for their outcomes (Deegan & Drake, 2006). It follows that shared decision-making should be incorporated in housing services so that clients are involved and feel responsible for their placements. Clients often did not know or understand housing options, including the different programs, funding, and housing types available to them. As a result, housing decisions were often left to the discretion of treatment providers who were privy to this information. One way to possibly remedy this is to have treatment providers conduct regular interviews with clients and family members where housing information can be mutually exchanged.
There are numerous areas for further study. This study focused exclusively on adults with dual diagnoses, and whether these results generalize to clients without a substance use disorder has yet to be determined. Further investigation is needed on recovery-related preferences and experimental approaches testing the utility of stage-wise housing placements should be explored. Another important question for future study is how to balance client preference with client need, particularly in light of some difficulties with insight (Goldfinger et al., 1999; Schutt & Goldfinger, 2000) . Also, given the variation in the kinds of supervised housing, studies should examine what constitutes the most effective structure for which client.
This study had several limitations, including limited data on the background characteristics of participants. Extensive residential histories were not gathered and reports of current housing experiences may be dependent upon past experiences, e.g., some clients may have never lived in independent housing before. Quantitative measures of stage of recovery would have also added another level of analysis. The basic demographics of the study sample are roughly similar to other studies of this population in Chicago (Rollins, O'Neill, Davis, & Devitt, 2005) and in national housing studies (Lipton, Siegel, Hannigan, Samuels, & Baker, 2000; Rosenheck, Kasprow, Frisman, & Liu-Mares, 2003), but the housing in this study may not be representative of housing in other areas. Nonetheless, our findings show the potential of housing programs and light the path for further questions and inquiry.