“Housing is Health Care”: Treating Homelessness in Safety-Net Hospitals

Abstract

As medicine integrates social and structural determinants into health care, some health workers redefine housing as medical treatment. This article discusses how health workers in two U.S. urban safety-net hospitals worked with patients without stable housing. We observed ethnographically how health workers helped patients seek housing in a sharply stratified housing economy. Analyzing in-depth interviews and observations, we show how health workers: (1) understood housing as health care and navigated limits of individual care in a structurally produced housing crisis; and (2) developed and enacted practices of biomedical and sociopolitical stabilization, including eligibilizing and data-tracking work. We discuss how health workers bridged individually focused techniques of clinical care with structural critiques of stratified housing economies despite contradictions in this approach. Finally, we analyze the implications of providers’ extension of medical stabilization into social, economic, and political realms, even as they remained caught in the structural dynamics they sought to address. [homelessness, housing, hospitals, marginalized populations, inequity]

“Housing Is Health Care,” read a button on a bag belonging to Jaimie,¹ a nurse in an urban U.S. safety-net hospital. Jamie and Sofia, a community health outreach worker, worked in a program supporting unhoused and underhoused patients. To stabilize patients’ health, they “treated” housing problems in addition to chronic diseases. At a biweekly case conference in their windowless office, Jamie and Sofia met with a supervisor to review caseloads. One by one, they discussed patients’ conditions, medications, whether they were showing up for appointments—and, importantly, their current housing status. Sofia explained:

The nature of working with this population is that a lot of stuff gets done, and then we need to stop and then wait things out. … [A] lot of sad things that happen between that. People end up back on the street while they’re waiting for transitional housing, or they’re kicked out of the shelter, or they’re on the senior housing waiting list for eight years. Even in [facilities], it’s not necessarily stable … because [maybe] it’s a prime location for gentrification.

Facilitating eligibility for scarce and dwindling low-income housing were key to stabilizing patients’ health and life conditions.

“Housing Is Health Care” is the unofficial motto of the National Health Care for the Homeless Council, following a previous claim made by AIDS activists (NHCHC and Lozier n.d., Fullilove 2010: 608). Safety-net health workers put this maxim into action by approaching housing as a therapeutic intervention within an affordable housing shortage. In this article, we consider how these health workers attended to housing-related issues and in so doing, generated openings and encountered limits in care practices that engage the “economic and political conditions that produce and racialize inequalities in health in the first place” (Hansen and Metzl 2014: 127). By examining providers’ efforts and struggles to confront structural problems in everyday hospital-based care work, we contribute an understanding of how frontline care providers navigate the contradictions involved in confronting a structurally produced dispossession within the individualizing paradigm of the biomedical apparatus.

We focused on multidisciplinary hospital-based health care teams providing specialized care coordination services for patients with low or no income and who are very ill and/or substance-using. We examined these “complex care management” teams in two U.S. urban safety-net hospitals.² Each team included registered nurses, physicians, social workers, and community health outreach workers (CHOWs). CHOWs, or “health coaches,” were most directly involved in service provision, and had comparatively low salaries. Teams coordinated care for patients with both medical risk factors (typically multiple chronic conditions) and significantly limited access to housing, food, transportation, or income. Health care workers saw lack of housing as a principal determinant of poor health. Finding sustainable shelter thus became one of the central foci of the team’s work alongside the work of helping patients access health care and manage chronic disease.

In the settings we researched, housing was increasingly defined as a direct and vital “treatment” for ill health (Kushel et al. 2006; Singh 2016). This follows shifts in care, social work, and policy addressing social needs—e.g., the “Housing First” models initially advanced by activists (Voronka 2019; Willse 2015). Health workers in this ethnographic study enacted housing as health care to stabilize patients’ lives as a precursor to stabilizing health. Yet this required health care workers to engage with highly commodified and stratified real estate economies, which have produced demographic shifts through rising housing prices accompanied by gentrification and displacement.

How did these health workers define and address patients’ lack of housing as a “health problem”? Drawing on ethnographic data, we show how health workers: (1) understood housing as health care and navigated the limitations of individual care provision within the structural reality of decreasing availability of affordable urban housing; and (2) developed and enacted practices of stabilization across both biomedical and sociopolitical registers. Finally, we discuss the implications of providers’ extension of medical stabilization into social, economic, and political realms, even as they remained caught in the structural dynamics they sought to address.

Background

Homelessness as a Health Problem

Researchers and policymakers have long debated the multiple origins and drivers of homelessness and have adopted differing explanatory models for homelessness, ascribing it varyingly to moral failure, pathological conditions, or political-–economic relations (Bassuk et al. 2014; Lee et al. 2010). Gowan (2010: xxi, 29) terms these “sin-talk, sick-talk, and system-talk,” and shows how each framework relates to distinct strategies for mitigation and prevention: criminalization (Herring et al. 2020; Lara-Millán 2014), medical treatment (Lyon-Callo 2000), and housing and social change (Gray-Garcia 2013; Matheiu 1993).

Writing in 2000, Lyon-Callo marked a partial turn toward medicalizing and away from criminalizing homelessness. Health care workers and institutions have therefore been increasingly expected to address homelessness and its associated problems. Lyon-Callo (2000: 330) identifies this shift as developing “a more effective and caring response to homelessness.” However, he also points out that clinical medicine is focused squarely on individualized treatment, and its interventions continue to marginalize solutions that “challeng[e] broader political economic processes” in favor of those focused on “reforming and governing the self” (Lyon-Callo 2000: 328).

As such, some scholars have questioned the utility of defining homelessness as a health problem. For Gowan (2010), “sin-talk” and “sick-talk” converge in their focus on self- governance, while “system-talk” more effectively looks to structural rather than individual causes of homelessness. Health researchers are also exploring mounting tensions between individual and structural perspectives on reducing homelessness (Luchenski et al. 2018). Mabhala and colleagues (2017) note that there has been a recent paradigmatic shift within health and medicine that recognizes “homelessness [as] a more complex social and public health phenomenon than the absence of a place to live.” These analyses converge with “structural competency” proponents who urge clinicians to “increas[e] recognition of the ways in which social and economic forces produce symptoms” (Hansen and Metzl 2014: 126). While this is an increasingly influential current in medical education and provision, there remains much to understand about how structural factors are understood and addressed in everyday practice.

Some scholars suggest that the very language of homelessness—including in health care provision and research—diverts attention from the central problem: the political economic distribution of housing. Willse (2015: 2), for example, discusses “housing deprivation” rather than “homelessness,” emphasizing that housing policies actively prevent some people from being housed.³ Rothstein (2017) asserts that the paucity of affordable housing in the United States can be attributed to the combined forces of the real estate industry, legacies of redlining and zoning laws, the attrition of public housing infrastructures, and the structural racism that shapes each of these. Fields and Raymond (2021: 5) employ Robinson’s (2000 [1983]) notion of “racial capitalism”— the contingency of capital accumulation on racialized devaluation—to describe how “racial hierarchies of value” shape the “accumulation by dispossession” that characterizes housing financialization. Scholars including Lyon-Callo (2000), Knight (2015), and Willse (2015) assert that such dynamics are obscured by notions of homelessness that locate the problem in the individual.

Yet theories of individual pathology presently form the basis of most health-based interventions (Wasserman and Clair 2014). Meyer and Schwartz (2000: 1190) note: “despite the conceptual understanding of the role of structural causes of homelessness, homelessness has been studied as if it were a disease, an outcome defined as residing in the individual.” Willse (2015: 54) calls the resulting treatment approaches “individual-reparative strategies.” He suggests instead that the racialized economies that unevenly distribute housing in fact produce the very illnesses and addictions that individual-reparative therapies aim to treat (Willse 2015: 55).

Housing Policy, Health, and Stabilization

Medical diagnoses—especially mental health and chronic conditions—remain among the few means through which people can attain sustained support (Hansen et al. 2014). Medicalized notions of homelessness are therefore imbricated with hospital work on a policy level. Eligibility for housing programs often depends on being connected with health workers who substantiate diagnoses and issue referrals (US DHHS 2014). As Knight (2015) discusses, qualifying for disability (SSI/SSDI) opens eligibility for certain housing programs (e.g., federal Shelter Plus Care or other state or municipal programs). Health workers thus became “gatekeeper[s] of resources” (Brodwin 2013: 21). This involves considerable effort of the part of providers as well as patients—a phenomenon that Darling (2016) calls the work of “eligibilizing.” Knight (2015:105) asserts that the “compensatory avenue” of diagnoses can help patients obtain “stabilizing economic and housing benefits,” but simultaneously consolidates the regulatory and punitive role of medical institutions in poor people’s lives. Some consider eligibilizing work part of providing good health care: van Eijk (2017) describes the “clinical-administrative work” of eligibilization in transgender health clinics as an “ad hoc corrective” to systematic failures of care coverage (van Eijk 2017). However, as Brodwin (2013: 22) describes, mental health providers working with patients in recurrent crisis also experience the “futility” of eligibilization in the milieu of extremely limited housing options.

One of medicine’s primary objectives is to stabilize patients: It intervenes on bodies that are out of sorts to restore them to a state of health. Panning out from the medical charge to stabilize individual patients, hospital-based health workers are increasingly foregrounding social stability through their growing emphasis on population health (Franz et al. 2016). One hospital described this as “viewing the neighborhood as the ‘patient’” (Kelleher et al. 2018: 1). Some public health scholars describe this as “sociomedical stabilization” (Connors et al. 2017: 229). This view of community health combines possibilities for improved population health outcomes with the potential for lower utilization rates and cost savings (Fleming et al 2017; Fleming 2019; Napoles et al 2017).

Hospital-based complex care management programs have arisen at this nexus of concerns about care and cost, which also drives hospitals’ financial interests in securing housing for patients (Fleming et al 2017; Kuehn 2019; Martin 2015; Powers et al 2020). Housing has become a vested interest for hospitals since 2013, when the Hospital Readmissions Reduction Program (HRRP) reduced Medicare payments to hospitals with comparatively high rates of readmission (CMS 2019). Readmission rates among unhoused patients are three times those of housed patients, so hospitals are increasingly interested in the net gains that housing their patients might produce (Saab et al. 2016).

Health and policy research on social determinants of health—“the social factors and physical conditions of the environment in which people are born, live, learn, play, work, and age”—has also encouraged hospitals to integrate “social needs” into care practices (Commission on Social Determinants of Health 2008; NASEM 2019). Others find that even this framework “takes for granted the larger historical and political framework … that determines the determinants” (Dubal et al. 2021). In the case of housing, “system”-focused scholars identify racialized political economy, accumulation, and disinvestment as destabilizing and depriving people of both housing and health care (Gowan 2010; Willse 2015; Wright and Rubin 1991). Once again referencing Robinson’s (2000 [1983]) understanding of enmeshed racialized devaluation and dispossession, Pirtle (2020) posits racial capitalism as a “fundamental cause” of health inequities. Willse (2015: 104) further argues that housing-focused medical interventions reproduce the systematic forms of racialized subordination that bring about housing deprivation in the first place. In these accounts, hospitals play a range of roles from ameliorative to damaging when it comes to racialized poverty.

As structural concerns become more recognizably consequential for health workers, new practices and dilemmas emerge that require more examination within medical anthropology. By studying the practice of “treating” housing in biomedical contexts, we demonstrate how health workers tacked back and forth between individual concerns and their interpretations of the broader problems at their roots. An expansive notion of stabilization—an ethnographically generated term—enabled health workers to enfold interventions like housing into patient care and assert a broader need for doing so within their institutions. We extended this concept theoretically to explore how work practices related to housing took up structural problems within patient care and inquire ethnographically into what phenomena and forces are stabilized in this work.

Data and Methods

This qualitative study is part of a larger multi-methods study seeking to understand factors contributing to effective care and retention for patients in safety-net hospitals and was approved by the Institutional Review Boards (IRBs) of the university and hospitals involved. We studied outpatient clinics in two U.S. urban safety-net hospitals, focusing on complex care management (CCM) teams providing interdisciplinary care to patients with multiple chronic conditions and concurrent socioeconomic challenges. CCM programs aimed to coordinate care and resources for chronically ill, marginalized patients to reduce expensive emergency room visits and hospitalizations.

Setting

The two hospitals studied differed in key ways around housing concerns, despite being located in the same state and region. Omega Hospital had dedicated teams of social workers, nurses, and health coaches supporting homeless or unstably housed patients. Omega Homeless Care worked with unhoused patients with multiple chronic conditions. Omega Underhoused Care was a team formed during our study that worked with people who were homeless or precariously housed and who used substances.³

Beta Hospital is in a city with a more robustly funded network of government and nonprofit housing services. Although Beta had a CCM team, they did not generally directly assist with housing. Instead, they referred patients to emergency social workers who supported them with housing needs and referred them to community agencies. While both hospitals employed health coaches, only Omega had a housing-focused health coach on staff.

To participate in CCM programs in both hospitals, patients needed to meet eligibility criteria. While criteria changed over time (Authors 2017, 2018), they usually required that patients have three or more hospital admissions in the prior year. Core CCM teams in both sites worked closely with patients’ primary and specialty care clinicians, including mental health providers and pharmacists. Health coaches stayed in touch via home visits and calls, provided health education, worked with patients’ families or friends to implement care plans, and “coached” patients to develop their capacity for self-management (Authors 2017).⁵ Social workers focused on short-term mental health support, connection with community resources, and relationships with family and friends. They also conducted needs assessments, contributed to care planning, and made recommendations to the team.

The Omega site is overrepresented in this article since it was part of the formal role of staff to address housing instability. However, concerns about housing permeated both sites.

Participants and Recruitment

Although this article focuses on staff members, participants included both patients and CCM staff members with a role in patients’ care. Two ethnographers at each site conducted observations by shadowing staff and patients and identifying eligible participants for interviews. Medical directors in each study site helped facilitate recruitment of program staff. Inclusion criteria required that participants be connected with CCM programs as a program staff member or care provider for current, potential, or past program enrolees. The research team also recruited health workers at community-based organizations with whom CCM programs partnered based on the same inclusion criteria.

Data Collection and Analysis

Ethnographers recorded interactions with patients and among staff to understand how non-physician health workers encountered problems, developed tools, and deliberated decisions. Four researchers shadowed staff and patients in clinics, homes, and other settings, conducting roughly 1,500 hours of observations between February 2015 and June 2017. During observations, researchers accompanied staff in their daily work settings and took detailed field notes. Weekly observations took place during case conferences, which were typically held among multidisciplinary staff teams in a hospital meeting room. Researchers also attended patient home visits with staff, which took place in temporary shelters, apartments, houses, SROs, outdoor locations, vehicles, and other dwellings. In addition, researchers observed staff in their offices as they interacted with patients, colleagues, pharmacies, and housing programs by phone and email. Field notes were anonymized to protect participant confidentiality. Ethnographic observations in this study served a critical role in facilitating interpretation of interviews, identifying key themes, and understanding the milieu within which health workers reflected on their work. In this article, we focus on qualitative interviews and observations with health workers. We conducted two provider/staff interviews with each participant. Participants were paid $25 per interview for their time. All interviews were audio-recorded and transcribed verbatim.

Transcripts were coded and analyzed using constructivist grounded theory (Charmaz 2014). Data analysis began immediately after collection and an initial review. Seven researchers developed a list of codes and subcodes. Each ethnographer coded his or her own notes and interview transcripts using ATLAS.ti. Queries involved both specific codes and relationships between codes, as well as data about frequency and other discernible patterns.

Participant Characteristics

The research team conducted 52 interviews with health workers (33 non-physicians). Most (77%) were female, 40% were White, 23% were Black, and the rest were non-Black people of color. Staff demographics differed significantly from those of the patient population, which was majority male (70%) and Black (63%), with most (83%) reporting having been homeless.⁶

Findings

In what follows, we examine housing as a central concern for health workers as they developed care practices to stabilize patients’ biomedical disorders and broader life conditions. We first describe health care workers’ views of housing economies in the region in which they worked, and how they worked to treat homelessness in this milieu. Second, we discuss how health workers treated homelessness by extending the notion of stabilization from the realm of biomedicine to political, economic, and social realms. This enabled them to define a lack of housing (not just homelessness) as a primary problem in which they intervened by devising care techniques through program eligibility and data tracking.

Care in a Housing Crisis

“It’s hard to manage your medicines if you sleep outside,” said Rhonda, a nurse at Omega, during an interview. She had been discussing the reasons why her patients needed additional support from the care team. In describing the goals of the program, Rhonda said:

Basically it’s to increase people’s functionality and to improve their housing. As a nurse, there’s very little [I can do] … because people’s health is so contingent on all the other areas of their life. Even though my goal is to improve their health … a lot of what I do isn’t about medication, it’s to try to improve their whole life and then that allows them to improve their health.

Rhonda and other health workers noted how the development or exacerbation of disease, injury, substance use, and mental illness were closely linked with patients’ lack of shelter. Describing one patient whose health was improving, having been housed in the course of the program, she said, “[I]f his housing situation changes he could be back on the street. … It’s just been so amazing. He was so ill [before he was housed]. He was [at the hospital] all the time.” Hence, she and other health workers discussed wanting to know and track patients’ housing status.

Health workers were especially concerned about the effects of changing real estate economies on patients’ health. In the study site as in other metropolitan areas in the United States, the costs of rental and real estate markets are soaring, related in large part to financialization and speculation in urban real estate markets (Fields and Raymond 2021). These shifts are linked to rising rates of foreclosure, forced displacement, homelessness, and racialized wealth inequity (Desmond and Shollenberger 2015; Fields and Raymond 2021). Our study sites have undergone marked growth in development, increases in housing costs, lingering effects of the foreclosure crisis, and a reduction in affordable housing.⁷

Kathy, a social worker at Omega, sighed during a case conference where the team discussed patients in a variety of living situations waiting for housing. Referring to the housing crisis, she said: “It’s hard. It takes longer for people to find housing.” Tina, a nurse at Beta, related personally to these circumstances, commenting that she was “priced out” of the city where the hospital was located. Her salary was likely about three times that of some of the non-licensed community health outreach workers with whom she worked—a class divide that was largely mirrored by the racial stratification of both teams. Health workers across pay grades frequently acknowledged the effects of the housing crisis in their own lives while speculating that the effects on their patients were even more acute. They generally described these dynamics through class inequity or immigration status, but rarely mentioned racialized inequities. Many health workers commented that it was unlikely that low-income patients could afford even cheap rent in the city. For patients who were currently unhoused, health workers described waiting lists for housing programs ranging from months to years.

In another case conference, Sofia described the housing situation in the region as “horrible, horrible, horrible.” She worked primarily with unhoused patients. While she and other community health outreach workers were under pressure from coworkers and clients to find housing, it was largely unavailable. She said:

I cannot stand having the conversation about … what’s available to them. I don’t wanna have to say “This is what’s available and in your price range.” I would love to be able to say, “Here’s a unit, here’s a deposit, here’s who you call for furniture,” then my job would be so much fun. … Honestly, it’s why a lot of people don’t want to be housing specialists anymore, because it’s depressing. … Landlords can simply wait for people who can afford to pay full price, so they’re not willing to rent to you.

During an interview with Eric, another health coach, Eric commented: “In our eyes, what success looks like for a homeless patient is not necessarily a house, because we know that it’s slim pickings right now.” Jaimie—the Omega nurse who worked primarily with unsheltered patients—agreed that it could sometimes be a “depressing job because there’s not affordable housing” in the area.

Health workers were also concerned about the insecurity and instability that patients experienced even if they were currently housed. They referred to the narrow margins by which people avoided loss of housing, and the profound consequences of even minor increases in costs. In an interview, Sofia commented:

I have seen a person who works really hard and is trying to manage their life with health issues, mental health issues, and has this income and they’re doing really good managing it, and then their rent increases. Sometimes as little as $50 and that has literally changed their life. They can’t afford it and there’s nowhere else for them to go.

She emphasized how this precariousness was compounded by illness, injury, and disability— especially when these affected people’s ability to work. Further, federal definitions of homelessness often excluded these patients from certain housing programs.⁸

Just as small negative shifts were accompanied by drastic effects for patients, Eric thought that incremental improvements in housing could also be extremely useful for patients, particularly given the lack of housing availability. During an interview he said: “If we can get you into [a single-room occupancy hotel], they let you stay for about 6 months … and then throughout all that you have a case manager who applies for housing with you, for different housing opportunities.”

While this might involve long waiting lists for permanent housing and perhaps multiple short-term moves, Eric described how for many patients, this could be “what success looks like for a patient that needs housing.” But he and other health workers also worried that expecting patients to move in and out of a series of temporary residences could destabilize their health.

In sum, it was not just patients without shelter who health workers focused on, but also those who were “underhoused.” Health workers were concerned not only with how a lack of shelter made people ill, but also with how illness could result in loss of precarious housing. Both housing instability and unstable health related to housing were key concerns for health workers vis-à-vis their understanding of regional housing economies, blurring the lines between “sick-talk” and “system-talk.” At this juncture, they frequently confronted the futility of confronting system problems through the apparatus of treating sickness.

Individual-reparative Techniques in Structures of Housing Deprivation

The individual focus of clinical work prevented health workers from intervening directly in the housing economies that produced rising rates of instability and housing deprivation. Nonetheless, health workers did what they could to address patients’ housing status as a primary health intervention. They developed new systems and reconceptualized their work to better incorporate housing as a pressing focus in patient care.

They observed that it was nearly impossible to address other health concerns in the absence of stable shelter, but rising housing instability meant more patients whose health problems they could not sufficiently address. Health workers also worried that standard health care practice failed to attend to or systematically track housing status, despite this being crucial to patients’ health. In addition, they expressed frustration when other health care workers did not prioritize housing as a critical intervention—e.g., when patients living without shelter were “discharged to nowhere,” even against hospital policy.⁹ Below, we describe the practices by which health workers tried to simultaneously stabilize housing and health care as paired care objectives. Two of the primary ways they accomplished this were through eligibilizing (Darling 2016; Hansen et al. 2014; Knight 2017) patients for housing programs and implementing “housing only” status into data-tracking practices.

Eligibilization

Health care workers identified their primary tasks as helping patients manage illnesses and secure housing. The safety-net health workers we observed enacted housing as health care in the everyday work of care delivery, but this was no small endeavor given the complex and shifting requirements of housing programs. Often, it required the labor of eligibilization—in short, knowing what resources were available and emphasizing the life conditions that made patients a good match for specific programs, despite the complexity of both tasks (Darling 2016).

This required that health workers be familiar with federal, state, and local housing programs and their ever-shifting eligibility requirements. The lack of a standard definition for housing instability also meant that Omega’s many unstably housed patients rarely qualified for housing programs for unhoused people, despite sharing similar health risks and concerns. Health workers became de facto housing advocates charged with, as Knight (2015: 117) puts it, “enabling social stability.” In her analysis of poor substance-using mothers, Knight (2015: 10, 117) describes “neurocrats” as new social actors that facilitate disability benefits for clients to “treat” poverty. Health workers’ tactics of eligibilization mirrored those of neurocrats: Members of the team did everything in their power to help patients obtain diagnoses and qualify for programs. They kept up on a wide range of medicalized and non-medicalized housing programs run by municipalities and private organizations as well as state and federal programs.

However, aligning programs and patients was difficult for health workers and often seemed fruitless. For example, Knight (2015) and Sufrin (2017) describe how substance use and histories of incarceration sharply narrow access to housing programs, and health workers struggled mightily to surmount these barriers. This obliged them to maintain a depth of knowledge about patients’ histories, program requirements, and potential paths to “legitimately” qualifying diagnoses. In a case conference, health workers brainstormed how to cobble together referrals and services to obtain a legitimizing disability diagnosis for a substance-using patient. Kathy said: “We don’t mind … linking them to services, but we also know it’s time-consuming and sometimes we get stuck.” Eligibilizing also involved difficult speculation: Jaimie described a patient living in his car who she was trying to enroll in hospice care to secure housing and end-of-life care. This required being six months from death. “He’s unpredictably close to death,” she said as the team debated whether to push this possibility.

Eric explained how Omega’s Homeless and Underhoused Programs worked in tandem to address both patients currently without shelter and those living in unstable situations. He described, for example, what often happened for patients who were ineligible for Omega Homeless Care:

A lot of times, they’ll see this person but it looks like they really need help, and so somebody from their team will say: …“They have substance abuse, they’re homeless, or they’re couch surfing; I think you guys [from Omega Underhoused Care] can jump in and do some good for him,’ and so we’ll bring him in, do a full assessment, and do an intake on him.

He asserted that health workers could easily predict that unstably housed people would decline in health and develop chronic illnesses. Patients who may qualify for Omega Underhoused Care, he explained, resembled what people in Omega Homeless Care “looked like a year ago, two years ago, two-and-a-half years ago.” He continued, “They’re usually that person who cannot make it to the primary care appointments … [for whom] it’s a constant battle between … food or … medicine.” The health workers we observed were confident that improving housing status—even prior to patients losing housing—could have profound effects on people’s ability to access consistent care and, importantly, would prevent their need to access Omega Homeless Services.

In some instances, eligibilization helped patients qualify for housing despite being in tension with cost-containment goals. During a case conference at Beta, health workers decided to refer an ailing patient to acute rehabilitation after hospital discharge primarily because it would qualify her for long-term supportive housing. They knew that readmitting the patient contravened the objective of reducing utilization but agreed that this was the patient’s only viable path to stable housing.

“Housing Only” and Data-tracking

Data-tracking practices were crucial for health workers as a means of demonstrating patient progress. One indicator of success was patient “graduation” or completion of the program, after which they were expected to use fewer hospital services. Some health workers expressed concern that having people linger too long in the program reflected poorly on its effectiveness, yet letting people go too soon could negatively affect patients’ “progress” and “self-management” (terms that emerged in case conferences).

Completing the program meant that patients’ risks for hospital readmission were not at a crisis point. In addition to standard charting, health workers tracked these risks though a parallel spreadsheet and care planning tool that rated patients’ stabilization needs across domains. In addition to medical goals and nutrition, these included income and housing—which standard charting afforded few systematized opportunities to track. Health workers “graduated” patients based on numerical ratings and team discussions. At one case conference early in the team’s formation, Bill, a physician supervisor at Omega, asked, “How do you feel about changing [the patient’s] status to ‘graduated’? There doesn’t seem like there is much more that needs to be done.” Becka, a nurse, replied, “My understanding is that graduation involves moving beyond all of our ancillary services, not just the nursing.” Bill replied, “[I] think about graduating as no longer needing our special services. I think he can get primary care and some counseling at a regular clinic.” Over time, however, the team experienced “graduated” patients cycling back into the program, which they observed was often related to problems with housing.

Omega eventually added a new classificatory category to their tracking: “housing only” status. This was for patients who otherwise met graduation criteria but lacked stable housing. Omega team members were particularly concerned about this group, since they observed that waiting lists for permanent housing were very long—years for some people. Because health workers thought housing was key to sustained health, they did not want to let go of patients before this need was met. However, in many cases, this significantly extended graduation timelines.

Sofia was hired as housing became a primary program focus, owing in part to her extensive previous experience as a housing advocate. She kept in touch with and supported housing only patients long after they were actively engaged in the CCM program. She accompanied them to meetings, communicated with caseworkers, visited prospective apartments with them, made referrals, and helped with applications. Patients sometimes used the team’s office as their address on forms or applications. One of our study’s ethnographers accompanied Sofia to a patient’s meeting at the Housing Authority. Health workers had helped the patient qualify for subsidized housing after a long period of unsheltered living. As the meeting was wrapping up, Sofia indicated that she would see him less often. He replied, “What, are you giving up on me?” She said no, but that that he had stabilized and would not be needing as much support. “Who said I’m stabilized!?” he retorted with a playful but anxious tone. She pointed out all the things that were going well for him, which he agreed were positive developments. He commented that his hospitalization had been oddly fortunate since it catalyzed these changes.

Other hospital staff reacted positively to the Omega Homeless Care team’s efforts to secure housing for patients, according to Sofia. In an interview, she said that the team was “making the medical staff more aware of the challenges of housing in the community that people live in [and] … helping doctors see that it’s not just curing the medical problem. …[I]t’s bigger than just a physical thing.” However, she added, “I think their expectations are a bit unrealistic, because they don’t understand the challenge of housing.” For example, Kathy explained that people could remain housing only patients in the program for “[a] long time. Months. … There’s people who are on our ‘housing only’ list that it’s now in the years.” Omega Homeless Care health workers described waiting lists for permanent housing that were up to four years long, and Kathy mentioned a woman who waited six years for senior housing. During a care conference, Omega providers discussed the idea of scaling up their data collection on housing hospital-wide by standardizing use of a “homeless diagnosis.”¹⁰ This did not gain traction, however, since there was little confidence that health workers outside the team would use it reliably. Furthermore, they knew from experience how difficult it was to keep current on patients’ housing situations.

Given these constraints, graduation from the program required a great deal of protracted work, as well as a willingness to compromise on housing conditions. Patients struggled to find sustainable permanent housing in the places where they had connections or roots. Furthermore, stratifying housing economies made it nearly impossible for health workers to feel confident that their patients would remain stably housed.

Extending Stabilization

Biomedical stabilization refers to a state of relative health, or at least an absence of embodied crisis. Stabilizing patients’ health was crucial for health workers across teams in each site, as stable health at least theoretically meant less hospital utilization. But for housing-involved health workers, simply intervening in physical health was insufficient. For them, stabilization went beyond biomedicine, extending into social, political, and economic domains. They extended this to include multitudinous forms of work, within and outside the domain of health, that aim to ameliorate deleterious conditions of life and health for people deprived of housing or other resources.

In an interview, Kathy described this as the ultimate objective of the CCM program:

[T]he goal is to support patients with stabilization. So, stabilizing their physical health conditions, their social issues that they’re having related primarily to housing, and then also to provide the mental health services as well to help them stabilize in that way, to address some of the symptoms that they are experiencing due to their current living situations.

For CCM staff, the possibility of stabilizing patients’ health required identifying and working on aspects of patients’ lives that preceded and precluded the possibility of biomedical stabilization. Jaimie commented:

Basically, no clinic in our system is set up for homeless patients. People who have no phone, who will no-show no matter what. No clinic- based intervention addresses those issues. We will show up [to where they live] in the morning. [Other hospital programs are] not set up to do that.

Conceptually and concretely, stabilization extended beyond hospital walls. Accessing safe and stable housing was key to stabilizing patients’ health, but housing economies presented a major barrier to accessing such stability.

During an interview, Eric conceded the limitations of individual advocacy but described how the team’s social stabilizing work “undoes neglect”:

[We serve] the downtrodden, the forgotten, the underserved … people in the community who are swept under the rug, and not really paid attention to. It’s really powerful being in a position where your job is to advocate for people who have, as a history, been neglected their whole lives, and that’s why they’re in the position that they’re in now, I believe, is because they’ve been neglected, their needs haven’t been met. To be in a position where we can actually address those things, and undo some of that, is definitely a worthwhile thing for me. … Change isn’t going to happen overnight, but on a person-by-person basis … it’s a great service to both the individual person, and the community at large.

While Eric was not explicit about the history of neglect to which he referred, it was clear that he saw his and the team’s role as remediating these broader conditions, its limits notwithstanding.

Sofia saw these issues as being beyond the team’s control, even as they constrained health workers’ ability to care well for patients. To address this impasse, she thought that hospital administration should provide low-cost housing directly to patients. During an interview, she said: “What … needs to happen is that developers that develop subsidized property, the city, and the Housing Authority [need] to partner with [the hospital] and they need to make affordable units available directly to patients.”

She was more interested in the hospital’s purchasing power than in institution-wide cost savings, which typically justify hospitals’ recent entry into housing markets (Doran et al. 2013). In the milieu of underfunded public housing systems, she saw safety-net hospitals as a potential force to stabilize structurally produced affordable housing shortages.

As such, Eric and Sofia agreed that “individual-reparative strategies” were limited but diverged in their estimation of whether and how these either contributed to or distracted from structural solutions to abandonment.

Discussion

As housing and its distribution (rather than individual homelessness) became a central concern for health care provision, health workers we observed found themselves increasingly implicated in housing economies, policies, and politics. Specifically, they identified a lack of affordable housing as a direct impediment to effective (and economical) care but struggled with their inability to directly meet this need. Instead, they worked to bring attention to the problem of housing, and improvised strategies to mitigate its effects on patients when possible. Regardless of their particular views on the efficacy of individual-reparative strategies, health workers focused on housing through expansive techniques of stabilization, through which they improvised care practices, leveraged evidence, and advocated for biomedical institutions to attend to inadequate housing.

It was in this sense that health workers enacted structural competency (Hansen and Metzl 2014). They ranged in the depth and extent to which they analyzed housing through an explicitly structural lens but frequently departed from medicalized views by invoking broader economic and political concerns as they discussed patients’ health. While rarely speaking explicitly of race or racism, they routinely referred to racialized neighborhoods, immigration status, and groups affected by inequity, displacement, neglect, and abandonment. Health workers were consistently frustrated by intensifying inequities in housing, as well as their perceived inability to influence them.

These frustrations led to experimentation and improvisation of novel care practices to extend stabilization through housing support. Livingston (2012) describes the role of improvisation in low-resource care settings. Her ethnography of an oncology ward in Botswana describes profound resource inadequacies, but also reveals the openings enabled by providers “making do, tinkering, and ad-libbing” (Livingston 2012: 21). In enacting housing as health care, health workers drew attention (albeit fleeting) to the ways that housing economies, racialized wealth distribution, and infrastructures—rather than an individual condition called “homelessness”—related to health.

For example, at one level, the housing only data classification reflected the limits of health care health workers’ ability to support patients’ health in the absence of affordable permanent housing. At another, it reflected their tenacious insistence that housing was a critical element of health, even as other health workers may have relegated housing status to the margins of care practice or defined homelessness as inalterably pathological. Further, in striving to reconceptualize stabilization as a social, economic, and political as well as biomedical process, health workers subtly asserted that biomedical care was ineffective or unsustainable in the absence of housing (among other resources). This worked in part to redefine homelessness as a problem of housing and economic distribution, and the problem of high utilization as an outgrowth of these circumstances.

Improvisation was also evident in health workers’ efforts to eligibilize patients. Like Knight’s (2015) neurocrats, they worked to translate social histories into medical diagnoses. Neurocrats, however, worked to diagnostically narrow social problems to enhance biomedical legibility. These health workers, in contrast, worked to expand the relevance of structural dispossession to biomedical practitioners. They pressed their colleagues to recognize historically shaped life conditions as crucial to the stability of patients’ health. By extending stabilization, they sought to convince others to reframe homelessness as a problem of distribution, disinvestment, or neglect and to transform infrastructures of care accordingly. In this regard, they worked implicitly to promulgate structural competency regimes even as they recognized the limits of their influence on these broader social problems.

Yet stabilization assumed the possibility of stability in an economic framework that routinely produces racialized crisis and scarcity. Drawing on Willse (2015), these sites of focused intervention rely on the very economic structures they seek to abate. For example, advocates of Housing First models were key in demedicalizing housing, but their expansion required making economic sense in a capitalist market. “[U]nexpected investments in vilified and long-abandoned populations,” he argues, came to “shore up and extend neoliberal economic industries that produce housing insecurity in the first place” (Willse 2015: 159). A host of related critiques have emerged around robust Housing First programs in Canada, revealing varied and new forms of racialized and economized governance (Fast and Cunningham 2018; Voronka 2019). Omega health workers reluctantly realized that housing-focused programs had to make economic sense to hospital administrators, and expanding beyond a small-scale, time-limited program would have failed in this regard. While their hopes for extending stabilization worked against the dynamics of racial capitalism, the program’s existence depended on its economic rationality within these very dynamics.

Likewise, Meyer and Schwartz (2000) argue that broad structural problems cannot be addressed solely within the domain of health care, as health-focused solutions are incapable of addressing the sources of inequities. Even the most optimistic health workers stopped short of viewing their work as structurally transformative. When asked about what Omega’s supportive housing programs would look like in an ideal world, Eric replied, “in an ideal world … this program would … not be needed … because everyone would have the basic needs that they need to survive and strive.” He then continued, “But we all know that’s not gonna happen.” Despite this inadequacy, health workers saw themselves as both improving care and aspirationally changing biomedicine’s ideas about housing deprivation.

This article contributes to emergent discussions about the openings and limits of structural competency in action. In addition, it raises questions for future research in this realm. For example, health workers engaged structural considerations at the level of housing markets, but these did not generally extend to a sustained historical view of racialized economic dispossession. To what extent can these distinctions in the depth and reach of structural analysis make a difference in practice? How might such differences shift landscapes of possibilities when it comes to both patient care and structural transformation? Such questions are ripe for further anthropological inquiry.

While the programs we observed did not altogether break from the “individual-reparative” approach against which Willse (2015) cautions, certain health workers—inconsistently and in contradictory ways—advanced distinct modes of understanding housing deprivation in relation to health. They extended stabilization to orient their practical work of confronting a structural crisis within an individualizing biomedical apparatus and to render these dynamics legible to others. Through their pragmatic elaboration of stabilization, they redefined patients’ primary problems, made do with the tools at hand, and built new ones to mitigate the worst outcomes for patients. In sum, this article offers an account of how health workers navigated structural competency in action, as well as the limitations they confronted in so doing.