Case Manager Perspectives on the Role of Treatment in Supportive Housing for People with Severe Mental Illness

Abstract

Objective

Growing recognition exists of housing as a social determinant of health, and thus, health care reform initiatives are expanding the reach of health care beyond traditional settings. One result of this expansion is increased Medicaid funds for supportive-housing programs for people with severe mental illnesses. This qualitative study explores the ways in which case managers working in a supportive housing program approach treatment and how their approach is influenced by both program requirements and their beliefs about mental illness.

Method

The study is part of a longitudinal qualitative study on recovery for people with severe mental illnesses living in supportive housing. Multiple interviews (n = 55) with 24 case managers from a residential-continuum supportive-housing program were conducted over 18 months. To provide an in-depth view of case manager perspectives, the study uses thematic analysis with multiple coders.

Results

Overall, case managers understand supportive housing as being a treatment program but predominantly characterize treatment as medication management. The following themes emerged: believing medication to be the key to success in the program, persuading residents to take medication, and questioning the utility of the program for residents who were not medication adherent.

Conclusions

Case managers understand supportive housing to be a treatment program; however, given the external constraints and their own beliefs about mental illness, case managers often equate treatment with taking medication. Study findings demonstrate the need to train case managers about mental health recovery and integrated health care. The findings also have implications for policies that tie housing to services.

By embracing the notion that housing is a key social determinant of health, the Patient Protection and Affordable Care Act (2010) has provided new impetus for the provision of housing for people with severe mental illnesses (Krieger & Higgins, 2002; Thiele, 2002). Much of this housing takes the form of supportive housing, defined as community-based programs that combine affordable housing with supportive services to help meet individuals’ health and psychosocial needs (O’Hara, 2007).

Housing provides many health benefits through primary prevention. People with adequate housing are less vulnerable to physical and mental disorders, chronic illnesses, injuries, and infectious diseases that result from living on the streets, in shelters, and in substandard or overcrowded housing (Krieger & Higgins, 2002). Secondary and tertiary prevention benefits accrue from housing by facilitating people’s access to health care and disease self-management (Henwood, Cabassa, Craig, & Padgett, 2013; Kidder, Wolitski, Campsmith, & Nakamura, 2007). Supportive housing builds on these benefits by adding a service component, which most commonly is case management. A team or an individual case manager work in the community with residents on a broad array of domains, including treatment and medication. Therefore, supportive housing acts as a health care intervention in two ways: through the housing itself; and through the provision of case management services, for which the housing serves as a site of health care delivery.

Background

Supportive Housing

More than 215,000 supportive housing units are funded through a combination of U.S. Department of Housing and Urban Development and Medicaid dollars (Solari, Cortes, & Brown, 2013). Typically, supportive housing programs move residents through a housing continuum ranging from more- to less-restrictive living environments, with residents’ progression based on treatment compliance and sobriety. The residential-continuum model is based on the clinical belief that people become housing ready by living in structured transitional environments that focus on treatment, abstinence, and development of skills for independence (Sylvestre, Nelson, Sabloff, & Peddle, 2007). To obtain a permanent home, residents are expected to demonstrate these skills as they transition incrementally through each level of housing; supervision is titrated downward and independence is increased as residents move closer to permanent, independent living. Over the last two decades, Housing First has reversed this model by placing people directly into permanent housing without requiring treatment compliance or abstinence. Although this approach is evidence-based, the Housing First model has yet to be consistently implemented (Padgett, Henwood, & Tsemberis, 2015; Rog et al., 2014). Moreover, supportive housing varies considerably depending on the local context, with differences in living arrangements, duration of housing, and program expectations regarding residents’ housing readiness and engagement in treatment (Siskind, Harris, Pirkis, & Whiteford, 2013).

Case managers are frontline providers working daily with people in their apartments and in the community on an array of issues, including medication, outpatient services, housing, employment, and social support. Either individual case managers or a case management team develop service plans with residents that create goals and prioritize tasks. To be reimbursed by Medicaid, service plans must tie goals and services to medical necessity criteria. Typically, case managers are para-professionals who receive ongoing clinical supervision to facilitate support and professional competency (Solomon, 2008). As frontline providers who tend to spend more time with residents than any other providers, case managers play a substantial role in how services are communicated and experienced by the resident. Case managers can be characterized as “street level bureaucrats,” working with large caseloads, ambiguous agency goals, and limited resources (Lipsky, 2010). To meet the accountability demands of the work, these providers often find ways to exercise discretionary authority. Particularly in the area of residents’ compliance and progression along the housing continuum, case managers make decisions about how they interpret and enforce program guidelines, and these decisions can profoundly influence the quality of services from the resident’s perspective.

Medication Management

Although they are not prescribing providers, case managers do work with consumers on medication management, and goal setting around medication often is included in their service plans. Medication management is implemented across health care programs and varies according to the role of the provider. The role of medication management is most clearly defined for case managers within Assertive Community Treatment, which is an intensive multidisciplinary team approach providing wrap-around services in the community. In this context, medication management is described as documenting the medication regimen and integrating that regimen into the treatment plan, connecting the resident with new medication, and documenting adherence (Allness & Knoedler, 2003). Case managers also are tasked with improving adherence to medication regimens by establishing a strong therapeutic alliance, being assertive in providing medication to residents, and reducing the complexity of medication regimens. Although case managers on an Assertive Community Treatment team rate medication management as the most beneficial mechanism of action, the focus on medication adherence has raised concerns among consumers about coercive practices (McGrew, Pescosolido, & Wright, 2003; Phillips et al., 2001). Within residential-continuum supportive housing programs, the role of medication management takes on an added dimension because compliance with medication and treatment is an important criterion in determining whether residents qualify for less-restrictive housing options.

Overall in mental health care, decision making about medication and treatment has shifted given the influence of the recovery movement (Deegan & Drake, 2006). The focus on recovery has been endorsed by the U.S. Department of Health and Human Services and is now a guiding framework for transformation of mental health services, moving care beyond symptom reduction to addressing what each person needs to pursue a meaningful life (Davidson, Rowe, Tondora, O’Connell, & Lawless, 2008). By reorienting care to personal life goals, the emphasis is less on pathology and more on building strengths, using both formal and informal supports, and allowing people to make decisions about their care (Tondora, Miller, Slade, & Davidson, 2014). Consumers’ perceptions of providers’ person-centeredness have been shown to predict recovery and quality of life for those receiving care in community mental health settings (Stanhope, Barrenger, Salzer, & Marcus, 2013).

Therefore, recovery provides a significant challenge to mental health care approaches that are focused solely on medication adherence and symptom reduction. With respect to medication management, Deegan and Drake (2006) have argued that the consumer should be considered an expert and that shared decision making should be based on a partnership that accounts for individual experiences, provider expertise, consumers’ right to autonomy, and information sharing. Shared decision making has been associated with increased satisfaction with services and improved social functioning among people with schizophrenia (Malm, Ivarsson, Allebeck, & Falloon, 2003). These values related to person-centeredness are now being promoted within health care reform initiatives (Alexander & Druss, 2012).

Health Care Reform

The Affordable Care Act offers opportunities to increase funding for supportive housing by providing financial incentives for health care agencies to partner with supportive housing programs. Under provisions of the Act, Medicare and Medicaid dollars can be redirected into housing through accountable care organizations, health homes, 1115 waivers (i.e., Section 1115 of the Social Security Act), and home- and community-based services (Alexander & Druss, 2012; Wilkins, Burt, & Mauch, 2012). The State of New York recently applied for a waiver to use Medicaid dollars to fund capital investment in housing, making the argument that supportive housing significantly lowers the health care costs of people who have high rates of utilization (Doran, Misa, & Shah, 2013). Many states are now implementing Medicaid health homes using a service delivery model that promotes care coordination for people with severe mental illnesses.

Health homes facilitate integration by creating a network of providers who develop one treatment plan and provide collaborative care across agencies through close communication and use of electronic health records (Alexander & Druss, 2012). Supportive-housing programs are encouraged to join health homes and collaborate with primary care providers because of the key role supportive-housing programs play in promoting individual wellness. Health homes are just one example of the new emphasis on integrating the delivery of primary care and behavioral health care. In response to the finding that people with severe mental illnesses die 25 years younger, on average, than people without severe mental illnesses (Colton & Manderscheid, 2006), all settings are now expected to provide person-centered care that attends to both physical and mental health needs. Therefore, the service plan must be individualized to the needs of consumers and relate to their personal life goals. In addition, the consumer must understand and consent to the service plan (Patient Centered Care Collaborative, 2012). This health care context is the environment within which supportive-housing programs are now operating as they move more explicitly into the health care arena.

As policies usher in the dual emphasis on recovery-oriented and integrated care, understanding how frontline provider beliefs and practices relate to service transformation provides valuable insights that can inform implementation. As part of a larger qualitative study examining recovery practices within supportive-housing programs, the current study focused on how case managers working in a residential continuum program provided care for people with severe mental illnesses. Qualitative methods provided the rich, in-depth data necessary to understand the nuances of how case managers regarded their work and how their work was shaped by their attitudes. We explored case managers’ attitudes regarding treatment and medication, the ways in which these attitudes influenced delivery of services, and the effects of program requirements on health care delivery.

Method

This study employed a longitudinal qualitative design to examine how frontline providers enact their case management roles with consumers living in supportive housing. A longitudinal design provided opportunities for prolonged engagement with providers, increasing the depth, rigor, and trustworthiness of data collection and analysis (Padgett, 2008). Sensitizing concepts of mental health recovery and recovery-oriented practice were used to analyze the data.

Sampling

The study setting was a supportive-housing organization that participated in a larger National Institute of Mental Health-funded qualitative study examining recovery among people with severe mental illnesses and histories of homelessness (Henwood, Padgett, & Tiderington, 2014; Tran Smith, Padgett, Choy-Brown, & Henwood, 2015). The primary intent of the study was to understand mental health recovery for the population of those with severe mental illness living in supportive housing. In addition to the resident sample, a purposive sample of providers was interviewed twice (at baseline and follow-up) to understand how services and service context influenced mental health recovery.

The setting provided supportive-housing services based on a residential-continuum model (Henwood et al., 2014). Case managers worked in two programs in a supportive-housing continuum: One program provided transitional housing, and the other provided permanent housing. Within this organization, the transitional-housing team had 16 case managers who worked with 87 service users; the average caseload was five to seven residents. The permanent-housing team had 18 case managers who worked with 370 residents, with an average caseload of 20 residents. In both programs, the majority of residents lived in shared apartments throughout New York City and received an array of services, including case management, socialization, support group services, and referrals for permanent housing placement and health care. The larger study inclusion criteria for consumers were a history of homelessness and co-occurring diagnosis of serious mental illness, and substance use. Prior to program entry, consumers were homeless (living on the streets or in shelters) and were referred to this program for housing.

The sample of 24 case managers was recruited for study participation based on their role as the primary service provider for those consumers sampled in the larger study. This case manager sample represented a 92% consent rate, with 2 of 26 case managers declining to participate. Two additional providers were not recruited because the resident-participant did not provide consent to interview them.

Case managers’ tasks adhered to an intensive case-management model, with case managers working with specific residents rather than in teams with a shared caseload. These tasks primarily focused on accessing, coordinating, and ensuring the receipt of services to assist residents, with a particular focus on maintaining housing stability. Table 1 shows the majority of the case manager sample identified as Black/African American (87%), having a bachelor’s degree or less (96%), and less than 3 years of employment at the agency (75%).

Table 1.

Participant Demographic Characteristics

Characteristic	All Participants (N = 24 (%))
Gender
Male	13 (54)
Female	11 (46)
Race/Ethnicity*
Black/African American	20 (87)
Latino	2 (9)
White	1 (4)
Education*
High School	3 (14)
Some Undergraduate	5 (23)
Undergraduate Degree	13 (59)
Graduate Degree	1 (4)
Employment Tenure*
Tenure < 1 year	8 (35)
Tenure 1–3 years	10 (43)
Tenure > 3 years	5 (22)
Program Type
Transitional housing	21 (88)
Permanent housing	3 (12)

^*Missing Data

Data Collection

The study protocols included multiple in-depth interviews with case managers over a period of 18 months. The baseline interviews with case managers occurred within a month of the resident’s enrollment in the study. There were up to two follow-up interviews at 6-month intervals or at the time of the resident’s discharge from the program. In total, the 24 providers participated in 55 interviews (34 baseline, 21 follow-up). Following the recruitment of resident-participants into the larger study, residents provided consent for the research team to contact the resident’s case manager for study participation. If case managers were no longer assigned to a particular resident at follow-up, no follow-up interview was completed.

The interviews were conducted by three trained qualitative interviewers familiar with the mental health service system and took place either in a private office at the case manager’s agency or at the study offices. The semistructured interviews ranged from 30 to 60 minutes and posed general questions about the case managers’ work and approach, as well as resident-specific questions. Questions focused on how case managers approached service planning and collaborated with residents on specific domains of the service plan. Case managers were asked about their beliefs about mental illness and the purpose of supportive housing.

Each interview was transcribed verbatim, and transcripts were entered into the qualitative software ATLAS.ti for data management. Interviews also collected relevant demographic information (e.g., education level, years of experience in the field). Following each interview, the interviewers completed an Interview Feedback Form, highlighting additional observations and areas for more discussion. These forms were shared with the research team, and interviewers engaged in peer-debriefing at weekly meetings. The authors’ university Institutional Review Board approved protocols for this study.

Analysis

Thematic analysis (Boyatzis, 1998) was used to generate category or template codes for similar quotes and ideas across and within transcripts, and ATLAS.ti was used to sort coded data. To develop an initial codebook, two researchers worked independently to complete open coding of 20 transcripts. Researchers then reviewed the remaining 35 (of 55) transcripts and revised codes to fit similar ideas across all transcripts. Codes were related to case managers’ discussion about residents’ insight into their mental illness, treatment and medication, and perceptions of a resident’s fit for the program. Coding inconsistencies were resolved through consensus.

The final codebook included codes that reflected case manager approach and practices related to medication, as well as descriptive, in-vivo, and process codes. These codes were then categorized into the three major themes: medication as the key to treatment success, working with residents’ toward medication adherence, and questioning the utility of the program for residents who were not medication adherent. The study design incorporated several strategies for rigor, including peer-debriefing, independent and co-coding, memo-writing, and prolonged engagement to aid in development of ideas and to provide a decisional audit trail (Padgett, 2008).

Findings

Overall, the case managers understood supportive housing to be a treatment program and, when referring to treatment, spoke mostly about their experiences working with residents on medication management. The primary theme that emerged was case managers’ stated belief that medication adherence was the key to success in the program. Within that theme were two subthemes: persuading residents to take medication, and questioning the utility of the program for residents who were not medication adherent.

Medication is the Key

Most case managers reported a belief that medication was the central mechanism for residents’ progress within the program, and therefore, medication was given priority over other potential interventions. Although case managers discussed the physical health of their residents, discussion around medication adherence nearly always centered on psychiatric medication. One case manager described medication as “the key,” and another stated that when people do not adhere to medication, “that messes with everything else in your life.” Case managers said they tried to have residents reflect back the same understanding of their mental health as that held by the case manager. In addition, case managers described their discussions with residents as focused on reinforcing the importance of the resident acknowledging his or her mental illness and need for medication. One case manager shared the following comment:

The challenge is to get them [residents] to the point that they feel comfortable with it [mental illness]. I mean, I said, “I have this diagnosis, number one. Number two, the medications are helping me to, you know, to maximize the reduction of my psychiatric symptoms, and to increase my functioning.” That level of understanding. “And, if I go into permanent housing, to keep my involvement with treatment, keep my doctor, get my prescription, even if it’s once a month, and keep taking my meds.” I think that’s important.

Case managers’ emphasis on the need for residents to develop personal insight often shaped their interactions with residents. The pressure to have residents endorse the same view held by the case manager was driven by a belief that insight was required for residents’ success within the system. The residents who were perceived to lack insight were often considered challenging. One case manager described that perception as follows:

Those ones that lack insight is probably one [of] the most challenging because I think in order for them to be completely stable, they have to be able to acknowledge that this is an issue that needs to be addressed. Like I said, the easy one is the one who has the insight, the one who is aware and knows that they can’t function without their treatment.

Case managers said they were aware that when residents were assessed for a less-restrictive level of housing they were far more likely to be successful if they expressed the view that they had a mental illness and should take medication. One way that case managers convinced residents to demonstrate insight was to ask residents to describe, in stark terms, the potential risks they perceived from not acknowledging their mental illness and taking medications. One case manager characterized the program in the following way: “This is a program to help you recognize that you do have an illness and you need to maintain it so you won’t go back in the hospital.”

Another case manager said he did not press the issue of medication with residents initially, but added, “Within a week you’re going to end up in the ambulance with the guy, and it’s not pretty.” Therefore, the case managers’ belief that medication was the key was not just driven by perception but by experiences of residents becoming symptomatic when they stopped taking medication. These experiences more often resulted in case managers’ invoking the threat of hospitalization as a way to persuade residents to become adherent rather than exploring why a resident might choose not to take medication. Case managers spoke about the need for residents to articulate side effects for monitoring purposes but rarely discussed how side effects could be a deterrent to taking medications. However, some said they recognized the struggles residents had with medication, including one case manager who stated,

He does take it as prescribed, he takes it every day, but he just doesn’t like that he’s on it, and it definitely makes him feel like, like he feels weak, you know, he feels like it’s a crutch, and it makes him feel bad that he has to take it and things like that.

When asked if residents would need to be on medications for the rest of their lives, most case managers said they believed that residents would need lifelong medication; therefore, the case managers’ goal was to set a pattern of adherence that residents could maintain long term. However, some case managers acknowledged that residents had differences in chronicity of symptoms: “For some, it [medication] might be for life. For some others, might not be for life. I’ve seen some people come off it, at a point, and they did well after that. It depends.” Even case managers who had a more nuanced view of the role of medication still understood their primary task to be convincing residents to take medication.

Monitoring and persuasion

The imperative around medication was not only driven by most case managers’ belief in its centrality to treatment but also driven by the concrete demands of their job. Because case managers worked with residents to graduate through the continuum of housing, the case managers were continually asked to document the extent to which residents were medication compliant. One of the most important aspects of documentation was the service plan, which was completed every 6 months with the resident and set the goals for treatment. Although some goals were selected by the resident, medication management was a non-negotiable goal in the service plan, which meant that residents and case managers had to be actively working toward medication adherence. As a result, new residents had to undergo intensive medication monitoring for at least a month. One case manager explained this process and its intended message to the resident:

The first 30 days initially is daily monitoring. Some counselors do go in and say “Let me see your medication” … They know they’re taking it but just want to make sure. And I think it kind of helps the client in a sense. It reminds them of how important it is if we are coming in every day asking to see it and asking them questions about it, it has to be important.

The monitoring sometimes took the form of counting pills, or case managers might request that residents take their medication while the case manager made a home visit. Case managers said they realized that this strategy was not foolproof; however, one case manager reported,

We only come by once a day, and at best we can watch them take their medication when they come see us. Most times they have like twice a day things, so it’s not like we can watch them all the time.

Attempts to verify whether a resident was taking his or her medication were intended to make clear that medication compliance was a priority for the program. Case managers frequently described medication management as a core activity—the one task they had to complete in their visits with residents even when it was not a priority for the residents. Case managers described the need to “cut to the chase” and check in about medication. The following comment by one case manager was typical of many:

Sometimes I go in and the residents don’t want to talk about the medication—“Let me talk about something else today.” I give them that option. And, at the end of the conversation, just give me 5 minutes. Lemme just look at your medication.

A major preoccupation for case managers was how to approach residents who did not endorse the beliefs around mental illness and medication as reinforced by the service planning process and program requirements. Case managers varied in the intensity and choice of methods to persuade residents to take medication. One case manager described his approach to this problem:

My approach is straightforward. “You can’t in this program” … in [program], and I think in [agency] as a whole is … we use the harm-reduction model—we can’t make the clients take their medication, but we can counsel. We can beat a dead horse.

This case manager’s approach was to discuss options with residents but ultimately allow the resident to make the decision. Moreover, this case manager alluded to having some flexibility that case managers could exercise to maintain people in housing. Although program requirements viewed medication compliance as necessary to progress in the program, case managers did not have the authority to force residents to take medication. This gray area gave case managers some discretion in how they pursued adherence. Some case managers described using more coercive tactics with residents who were not medication adherent:

You can’t be on their side in that way. You have to say, “Look, sir, you’re here. This is a program for mentally ill people. I need you to take your medication right now or we’re going to have to call the ambulance.”

One case manager, when asked if she received training on medication management, replied, “No training in how to get the clients to take their meds.” Case managers described the escalation that occurred when they perceived the resident to be nonadherent, which involved increased monitoring, counseling the client, consulting agency supervisors, having a case conference, and transfer to another program for residents who remained nonadherent.

“This isn’t the place for you.”

The transfer to a more-restrictive housing program was characterized by case managers as “a better fit” for residents. Case managers tended to reconcile themselves to the fact that some residents could not keep their housing by perceiving them as inappropriate for the program. This perception was predicated on a strong belief that the program was not primarily housing but rather primarily a treatment program that offered housing. Case managers frequently characterized supportive housing as a treatment program and described their efforts to convey this concept to residents:

She admitted to me, she’s like, you know, “If I didn’t need a place to stay I wouldn’t be in the program.” And I was like, “You know what, that’s fair enough, I understand the need for housing, but this is an apartment treatment program, so if you don’t want the treatment part, it makes it really hard, cause then when we ask you what goals you wanna work on, you don’t wanna work on anything!”

Residents’ acceptance of the supportive-housing program as a treatment program was signified to case managers by residents taking their medication. If residents subscribed to the belief that supportive housing was treatment, then the logical conclusion was that they should participate in treatment, which was mostly understood in terms of taking medication. Case managers expressed frustration with residents who were not adherent to medication and considered these residents inappropriate for their program:

Because we can’t force—you know, no one can force anyone to take any medication and, um, if it becomes a problem where their noncompliance is, um, is affecting their housing then we have to find a way to get them out of here. We have to find them a different level of care, because obviously they don’t think that they have a problem and this isn’t the place for them.

Case managers underscored the idea of housing readiness by articulating specific expectations that would allow residents to graduate to less-restrictive housing. Case managers most often assessed housing-readiness in these terms:

For those people [who] believe that they have no mental illness, it really doesn’t work for them. Because if they feel like they don’t have a problem, they’re not gonna take their medication, and they’re not gonna comply with any of the program guidelines … some people … they’re not ready.

Case managers often bluntly expressed the consequences of not being housing ready. Typical of this viewpoint was the following case manager’s appraisal of residents who were not taking medication or going to outpatient treatment: “Those people need to go.” Others articulated the challenge in terms of finding the right level of care:

Yeah, there are some clients that are so advanced that they shouldn’t be at this level. They should be at the next level. And … it’s just not doing them any good. And then you’ve got some clients that just, they’re not advanced at this level, they need so much monitoring. They need 24-hour monitoring.

As a consequence, case managers gauged the appropriateness of the program on the extent to which residents could articulate their view of treatment and demonstrate behaviors that reflected that view.

Discussion

The findings of this study illustrated how case managers understood adherence to medication to be key to residents’ success within the program. The majority of case managers in this residential-continuum supportive-housing setting understood their program to be a treatment program. Therefore, the desired trajectory for each resident was to acknowledge that he or she had a mental illness, and thus, to engage in treatment, which was primarily equated with taking medication. The themes of monitoring and persuasion show how the program’s prioritization of medication as treatment led to a focus on monitoring and persuading residents to take medication. Case managers were skeptical that the program was an appropriate fit for residents who resisted taking medication or for residents who indicated their only interest in the program was as a source of housing.

Although case managers worked on a variety of tasks with residents, the fact that medication adherence was considered the clearest indicator of treatment compliance led to its prioritization. Therefore, case managers perceived treatment as a singular, linear path to success through insight and medication. Their working assumption, “If you know that you have a mental illness, then you take medication,” led case managers to conclude that “If you do not take medication, then you must not understand that you have a mental illness.” Mostly absent was a focus on the resident’s perspective with regard to medication or a sense that taking medication was a shared decision.

The strong emphasis on medication adherence and the case managers’ strategies to achieve adherence with residents were determined by multiple factors. In terms of how residential-continuum supportive programs were structured, case managers were pressured to move residents to the next level of housing, which often required documenting resident compliance with medication. In addition, service plans that respond to the Medicaid criteria of establishing medical necessity have been interpreted as requiring active medication goals. In terms of their frontline practice with residents, case managers had neither the time nor the training to negotiate the complex area of medication adherence. With large caseloads and productivity requirements, case managers often spent as little as 15 minutes with residents, allowing little time for a case manager to explore the residents’ attitudes about medication or to build trusting relationships that would enable these difficult conversations. However, this approach to medication also appeared to concur largely with case managers’ beliefs, understanding, and experiences of mental illness. With their particular treatment lens, case managers were susceptible to the confirmation bias that all paths led to medication. Their experiences with some residents seemed to further solidify their thinking that all residents needed medication to move forward rather than an individualized approach to resident needs and preferences.

The findings from this study reveal practices that run counter to those promoted by mental health recovery and integrated health care approaches. Both approaches have championed person-centeredness over pathology and symptom reduction, generating a more holistic approach to well-being. Person-centeredness has stressed the role of individual strengths, living in the community and natural supports in managing wellness (Tondora et al., 2014). In contrast, case managers tended to adhere to a medical-model approach, which focused on the need to control symptoms through biological intervention. The issue of adherence to psychiatric medication has vexed the mental health system and created a significant tension between service users and providers for decades (Corrigan et al., 2012). However, the complexity of adherence given side effects, interaction effects, and the powerful stigma of taking psychiatric medication was not acknowledged by most case managers in this study. Pressured by the need to move residents through the housing continuum, case managers tended to simplify their approach to medication in ways that undermined residents’ self-determination. It was at these times that we most clearly saw case managers interpret program guidelines to meet the demanding requirements of their jobs. Whereas other health care sectors have been developing tools to promote decision support (O’Connor, Légaré, & Stacey, 2003), these types of tools are just beginning to be implemented among case managers (Woltmann, Wilkniss, Teachout, McHugo, & Drake, 2011). With regard to integrated health, case managers did not reference medication for physical disorders nearly as often or in the same manner as psychiatric medication, suggesting a priority in behavioral improvement over physical health.

Housing First approaches have made some advances in addressing these issues by endorsing housing as a right that is not contingent on treatment participation or sobriety (Tsemberis, Gulcur, & Nakae, 2004). Although Housing First case managers have more room to honor resident choice, they still need more training in shared decision making and decision support (Woltmann et al., 2011). The demands of Medicaid billing also persist, requiring case managers to complete service plans that fit within a diagnostic rubric (Clossey, Rowlett, & Walker, 2014). Case managers usually have the most contact with consumers but often have the least training among those who provide care. As reflected in this sample, the majority of case managers did not have master’s level clinical training, and most received minimal orientation to the work when they started their employment. Given that health care reform is expanding the view of health care to include a range of community based programs, the training and preparation of that workforce is vital if care is to be delivered in a way that aligns with the principles of mental health recovery and integrated health.

Limitations

Although qualitative studies are not designed to be generalizable to wider populations, the findings from such research might be transferrable to other settings (Lincoln & Guba, 1985). A limitation of this study was its restriction to one residential-continuum setting, and therefore, some of the circumstances facing the case managers might have been unique to that setting. Additionally, within that setting, several case managers were interviewed multiple times about their work with different residents, whereas other case managers were interviewed only once. We monitored the data in the analysis to ensure that the views of those who had been interviewed multiple times were not overrepresented in these findings. Finally, this study was informed only by frontline providers even though the constraints within which they worked were driven by program policies and procedures related to housing and health care. Therefore, including the perspectives of agency leaders and policymakers would give more context to case manager practices. Future research should include multiple perspectives to further understand how health care is delivered in supportive-housing settings.

Implications

In describing the State of New York’s efforts to redirect Medicaid dollars into supportive housing, Doran and colleagues (2013)s concluded, “For many patients, a prescription for housing or food is the most powerful one that a physician could write, with health effects far exceeding those of most medications” (p. 2376), endorsing the notion that housing’s primary contribution to health care is as a structural intervention. Moreover, for people with severe mental illnesses, housing’s contribution to health is borne out by the evidence, with Frank and Glied (2006) concluding that since deinstitutionalization, mainstream benefits such as housing have contributed more to improving people’s mental health than did treatment. Therefore, although the rhetoric appears to support the promotion of housing as a preventive approach, the reality is that new Medicaid dollars will likely go to programs in which the service component still relies on a medical-model approach. Moreover, residents who do not comply risk losing their access to housing, whereas many others in need of housing refuse to engage given the stigma of enrolling in a mental health program (Luhrmann, 2007). The overall lack of affordable housing options for people who need housing but who do not want to engage in supportive housing results in many people with severe mental illnesses ending up back on the streets or in shelters.

One way to address the problem is to take seriously the notion that supportive housing is a health care delivery site, which means ensuring that case managers are trained as health care professionals. The same innovations being encouraged in other outpatient health care settings could be incorporated into supportive housing, especially providing coordinated care for both physical and behavioral health needs and training providers in decision support and person-centered care. A more radical solution is to sever the ties between diagnosis and housing, as is the case in other peer countries to the United States, such as Denmark and the United Kingdom (Braga & Palvarini, 2013). This approach acknowledges the health benefits of stable housing but does not make access to housing contingent on diagnosis or treatment, but rather on need in the broadest sense. With diminishing welfare benefits and increasing poverty, social work researchers might argue that society has seen health care forced to step into these sectors for treatment to have any chance of success—and that the Affordable Care Act represents a significant expansion of this approach. Ironically, this strategy undermines the powerful health care benefit of housing that should be available to all people, regardless of diagnosis and health care-seeking behavior.