The effect of a non-congregate, integrated care shelter on health: A qualitative study

This qualitative descriptive study used individual semi-structured interviews to achieve its purpose. This study is part of a larger mixed-methods investigation exploring the health outcomes of people living in an ICS during COVID-19. The study began 18 months after the Haven opened. The participants were recruited from The Haven, set up in September 2020 as described above and elsewhere in the literature ( Rosecrans et al., 2022a ; Rosecrans et al., 2022b ). From May 2020 to June 2022, approximately 170 residents were served in the Haven. We used purposive sampling to ensure our sample included people that differed by age, race, and length of stay in the shelter. The patients were contacted by phone or in person by a research assistant who also worked as a residential staff member at the Haven. We included participants if they (1) were over 18 years or older, (2) lived at the Haven, (3) had a history of homelessness, and (4) were willing and able to provide informed consent. Participants were considered ineligible if they: (1) were pregnant at the time of screening, or (2) did not have the mental capacity to complete an in-depth research interview (as determined by a brief verbal consent quiz to confirm understanding of the study). The sample size of 20 participants was determined by referring to prior qualitative methodology literature, which indicates that this sample size is typically sufficient to achieve data saturation for analyzing major themes ( Braun & Clarke, 2012 ; Lindgren et al., 2020 ). Data were collected during face-to-face interviews while participants were in their private rooms at the Haven, using video online conferencing software (Zoom). Interviews were audio recorded using a recording device separate from Zoom. The interviews were performed from February to March 2022. The interviewer was not known to the participants before the interview to pursue an unbiased approach. The semi-structured interview guide used open-ended questions which invited participants to relate their experiences of homelessness before coming to the ICS, their experience at the Haven, and what they look toward in the future. The interview guide was developed after a review of the literature, and consultation with the Haven and research staff. Table 1 lists the interview guide questions. To record non-verbal communication and emotion, field notes were taken and analyzed with the interview transcripts, congruent with qualitative thematic analysis methodology. Participants were paid $40 USD on completion of one interview. WCG Institutional Review Board approved the study. All participants provided written consent. Verbal consent was additionally provided before the interview began, ensuring participants knew the project’s purpose and that the interview was audio recorded. Participants were assured they could pause or withdraw from the interview for any reason. No participants withdrew from the study. The research assistant was an employee at the shelter who could link participants to support and resources if they became upset during an interview. We protected the identity of participants within this paper, by redacting any proper names and using pseudonyms and descriptors only. After the interviews were completed, audio recordings were transcribed verbatim. Using NVivo software, each transcript was reviewed, and data were initially summarized for key points and frequently occurring statements. The research team then iteratively analyzed and categorized the data using Braun and Clarke’s six stages of thematic analysis: (1) familiarization, (2) generating initial codes, (3) searching for themes, (4) reviewing themes by re-reading, (5) defining the essence of each theme by developing three main themes, and (6) writing up the results. NVivo was used to organize the data into codes and then larger themes and to share the coded data with the whole research team. COREQ guidelines are used throughout the data reporting to strengthen the quality ( Booth et al., 2014 ). Data analysis occurred in May and June 2022. We used the criteria of credibility, confirmability, and transferability to increase the rigor of this work ( Fereday & Muir-Cochrane, 2006 ). The first author coded sections of the first transcript independently and then compared their findings to that of the three other authors to increase credibility and confirmability. We recorded all initial interpretation decisions as we progressed through analyses to final themes. Examples of transcript quotes and their relationship to codes, subthemes, and themes are seen in Table 2 . All authors participated in a discussion of the final themes. Transferability was improved by describing the specific setting details of the study, which put the findings in context.

We interviewed 20 people, 6 women, and 14 men, aged 23–71 ( M = 50). Participants’ demographics are described further in Table 3 . Medical co-morbidities and substance use detail were collected at baseline as part of the larger study. Physical and mental health co-morbidities ranged from depression and anxiety to brain tumors and epilepsy. Only one participant was actively in treatment for an ongoing substance use problem, although all reported a history of substance use. The most frequently reported substance used was tobacco ( n = 16, 80%), followed by marijuana ( n = 12, 60%), and alcohol ( n = 10, 50%). Six participants (30%) reported using cocaine or heroin. The interviews showed overall high satisfaction with the ICS model used at the Haven. Participants reported engagement with case management and nursing services, which significantly improved their perceived health status and positive, respectful relationships with residential and administrative staff. Three themes emerged related to the participants’ experiences of the ICS model at the Haven: Autonomy, Supportive Environments, and Stability/Need for Permanent Housing. The non-congregate space of the Haven allowed participants to regain their sense of autonomy. Participants described that having a private and secure room with respectful staff, was in a sense, “re-humanizing,” after frequent experiences in congregate settings. The disrespect people experienced in other settings came from staff and other residents and indirectly from having to live in a congregate setting. Participants frequently compared the differences between congregate and non-congregate settings and described how their mental and physical health improved in a non-congregate space. Participants reported that the non-congregate shelter model allowed an enhanced sense of security, autonomy, and support for medical needs that many residents with a previous history of living in congregate shelter settings found more appealing. I just know this experience right here is by far the best. I don’t feel like I’m out of place. I feel good, I feel safe here. I don’t feel like I’m homeless. [Robert] I mean, I do feel like they [Haven and hotel staff] try to make you feel like a human. You know, you don’t feel like a homeless person when you stay here, you feel like any other guest that walks in and comes in off the street. [Carol] Tanya, who used a wheelchair, had previously been in a motel-type shelter but felt that her physical needs were not met in that environment. Moving to the Haven was helpful in terms of increasing her physical mobility and aiding her activities of daily living. She reported that the Haven staff worked with the Lord Baltimore hotel staff to design appropriate rooms for people with mobility issues, including installing no-slip grips, handrails, and chairs in the bathroom. She said that she was checked on routinely, and her room was cleaned frequently to ensure clear pathways to the bed, door, and bathroom. Sandy reported that although she had previously lived in a motel-based shelter, they had been required to leave their doors unlocked. Someone had violated her personal space and safety in that prior shelter, and she no longer felt safe enough to sleep. Phil mentioned that the privacy of having one’s own bathroom, as opposed to a dormitory-style group shower, made for a better experience at the Haven. Shirley stated that her prior shelter had 30 women in one room. Joel enjoyed the ability to be alone at the Haven if he felt overwhelmed, I came from a place where there was like eight people in a room and there was no privacy. Just no recreation whatsoever as far as if I need to unplug because I’m overwhelmed. I can turn on the TV and watch TV [at the Haven]. Or sit here and read a book and it’s stone cold quiet or listen to music. [Joel] Physical and mental health were discussed interchangeably as important primary concerns. Substance use was discussed as either a past problem or a secondary issue. Multiple participants reported that their substance use was only a problem when stressed or when with other people who used drugs. Participants reported that having their own space, away from others, was beneficial to stress management. Participants detailed substantive informal and formal support received from residential staff, including security and front desk staff, and from case managers and nurses, resulting from multidisciplinary involvement and commitment. Participants felt they had a choice in their food, which was perceived as healthy and abundant. Matt reported that the prior shelter where he stayed only served cold, brown-bag meals. Participants also reported being able to obtain snacks and outside food when they wanted it. Kyle mentioned that the healthy food at the Haven was the main reason he felt his health had improved. Participants described positive experiences of the case managers working with participants to obtain identity documents, state-issued unemployment and disability benefits, transportation tokens/passes, and food stamp vouchers. Participants also detailed the assistance case managers gave them in their search for permanent housing (including placement in assisted living facilities due to participants’ high medical acuity). Three participants required disability-compliant housing, and participants reported that case managers were involved in identifying and touring potential housing units with Haven residents. In addition to the improved physical shelter environment, transportation and food were frequently identified as quality services offered at the Haven. If I’ve got questions, she’ll respond right away. Sometimes it’s as trivial as having her scan some documents and email them to me so that I can get them out to a medical provider or to social security, or unemployment. [Joel] Case managers were highlighted for their emotional support. When asked whom they would turn to if they felt overwhelmed, depressed, or anxious, many participants reported turning to their case manager. Sometimes, in the dialogue of an interview, they would refer to their case manager as a counselor. This place makes me feel comfortable and they’re not acting neither. They’re genuine and yeah, from day one, they’ve been very welcoming to me. [Kyle] Participants described their experience with the on-site nurse case managers, supported by nurse practitioners, who provided a variety of services, including assessment and triage of urgent clinical needs and referral to the hospital as needed, transitional care coordination after acute hospitalization, prescription refill assistance and medication adherence support, and outpatient care coordination. Participants also described the nurses as providing significant medication adherence support (including delivery of medications), and health system navigation. I have epilepsy and I suffered depression, which was determined way before the pandemic, before I got evicted and everything. So, this place, yeah it could have been very easy to go off the wall, not have the ability to provide for yourself and not have the ability to get that all-important seizure pill. [Kyle] Robert reported that everything had changed within months at the Haven, and he now had all his identity documents and a housing voucher. Matt had experienced nursing care at a prior shelter, but the prior shelter did not provide the services the Haven nurses provided. And they act like they [previous shelter nurses] didn’t care. I mean, they did just the bare minimum. But then, here at the Haven, everything’s A1. The service, the people. I mean, it’s like being in a nursing home. Everybody treats you with respect. [Matt] The location of the Haven also served as an important difference compared to other shelters, which impacted participants’ health. Transportation was accessible in the Haven’s downtown urban location, which improved access to medical appointments. Only two participants reported not having a primary care physician, and many reported seeing multiple specialists regularly, including therapists and psychiatrists. The participants felt that the Haven had improved their perceived health status. If my health wasn’t improved, there’s no way I could survive homelessness at all. [Kyle] The Haven provided many participants a sense of being “at home,” and a feeling of constancy in a life that had been heretofore tumultuous and changing. Routines were established, autonomy was secured, and identity was built or rebuilt over time. That’s what the Haven has been. They’ve been trusting resource for me to build a solid foundation. [Nigel] Although participants perceived support as improved at the Haven compared to other shelters, barriers to support and full independence remained. For example, participants reported that they were not entirely free from surveillance. Front desk staff and security still restricted the use of communal spaces and instituted nightly curfews (with exceptions for work and visiting family). Yet, participants reported having more control over their space and body than in previous shelter environments. On average, participants lived at the Haven for under a year ( M = 311 days; range = 74–536 days). Participants reported that the Haven staff collaborated to facilitate their stability in several ways. Haven staff protected residents’ privacy, respected their preferences in food, and worked collaboratively with patients to address health needs. In meeting many basic needs, foremost of which was the private sleeping space, the Haven enabled participants to focus their time and energy on their health needs and future, including permanent housing. At the Haven, participants had space to rest and reduce stress, which enabled long-term planning and care, including addressing health issues, completing paperwork for social service programs, and searching for housing. Participants overwhelmingly identified obtaining permanent housing as their primary need. Ken mentioned that his criminal background check prevented him from finding permanent housing. He also felt he would need to give up his freedom of choice to find housing. Donna mentioned that her housing options might be limited to places where she needs to climb stairs, which she could not do because of an injured hip. She thought she might be on the street if the Haven closed and forced to return to the shelter. Because like I told my caseworker, I cannot go in these shelters where you got to leave 6:00 in the morning and come back. I can’t walk very far or long without holding on. I just can’t do it. And that’s what I’m afraid of. If I can’t find something, that’s the only shelter available. [Donna] Regarding future unmet needs, the impending closure of the Haven was announced during the 11th interview and was mentioned by five participants as a significant concern. Participants expressed concerns, distress, and anger over the pending closure. Those participants who had lived at the Haven for over a year also reported changes in the Haven staff and environment leading up to the closure announcement. There was also a sense of apathy and resigned awareness of life’s instability in the way participants responded to the news of the closure. You don’t expect [good things] to last forever. Yeah, I wish it could last longer, but you got to move on. Yeah, so I don’t have no complaints about it. [Betsy]

During the COVID-19 pandemic, Baltimore City implemented two services co-located at a downtown hotel, operationally known as the Lord Baltimore Triage, Respite, and Isolation Center( Rosecrans et al., 2022a ; Rosecrans et al., 2022b ). The center included a medically super-vised COVID-19 isolation and quarantine service, and an integrated care shelter (ICS), known as “The Haven.” Residents were accepted to the Haven by completing isolation and quarantine at the Lord Baltimore Triage, Respite, and Isolation Center, or through transfer from another city shelter to expand temporary shelter capacity. Leadership and staffing for the isolation, quarantine, and Haven services were integrated. The University of Maryland Medical System clinical staff provided Haven’s clinical staffing with oversight from the health department clinical leadership. Haven residents stayed in individual hotel rooms, equipped with a personal shower, bathroom, bed, and closet space. This non-congregate setting contrasts with congregate shelter settings with more than 30 people in one room. Haven housing case managers assisted residents through the necessary steps to secure housing through the housing voucher process or other low-income housing options. Nurses assisted with linkage to medical services, care coordination, prescription management, and assessment and triage of urgent medical issues. Residential staff fostered positive relationships with participants and provided emotional support. Residential support and security staff provide 24/7 presence on the floor to advocate for residents and address immediate needs. To our knowledge, this is one of the first temporary shelters to adopt an integrated care model. While implementing the integrated care model represents an innovative strategy to meet the needs of a highly vulnerable population, information is lacking regarding the specific components that are acceptable and utilized by people experiencing homelessness in temporary shelters to improve their health. Thus, the first crucial step is to investigate the use of care components within an ICS model in a temporary shelter setting, and how residents perceive their health outcomes. The knowledge of a person’s lived experience within a temporary ICS and how this environment affects physical and mental health and well-being is largely unknown. Our study will add knowledge about the link between an ICS model in temporary shelters and its relationship to physical and mental health. This study aims to describe peoples’ experiences receiving care in an ICS setting established during the COVID-19 pandemic to meet the complex needs of medically vulnerable individuals experiencing homelessness or housing insecurity. Health is defined, for the purposes of this paper, as inclusive of physical and mental health needs.

Participants in this study were able to connect the ICS model with positive changes in their lives and health. Important components of the model included: (1) non-congregate space offering privacy, dignity, and safety and (2) on-site clinical support and care coordination. Despite these elements, a key need that impacted residents’ mental health was uncertain access to permanent housing options and instability of shelter services as the Haven closed. The novel non-congregate, ICS model and supportive service philosophy of the Haven may hold lessons for designing services for medically vulnerable individuals experiencing homelessness and housing instability.

This paper explored residents’ experiences in an ICS model within a novel, non-congregate setting established during the COVID-19 pandemic in an urban city in the United States and the model’s importance to health. A public-private partnership allowed for integrating housing case management and clinical services to provide a novel temporary shelter program that attempted to address the root causes of homelessness. Participants prior experiences in navigating shelters and medical care in the community noted that individualized, personalized care is not the norm. We identified ways in which living at the Haven and receiving integrated care services were perceived as important to whole-person wellness. Homelessness, living in unstable housing, and trying to find housing can be isolating experiences, despite sometimes necessitating living unwillingly in congregate settings. In the face of housing insecurity, participants spoke of a profound lack of autonomy. Research has highlighted several factors contributing to an unwillingness to use traditional, congregate shelter systems among people experiencing homelessness, including shelter policies that result in a lack of personal space and separation from one’s partner or family ( Moffa et al., 2019 ). Participants at the Haven explained the role of non-congregate shelters in terms of their autonomy, dignity, and safety, including reflections on how their experiences at the Haven differed from their prior housing situation. Participants described substantive ways the non-congregate setting supported their health by providing a less chaotic environment and a place to rest, which would not have been possible at a traditional congregate shelter. The non-congregate space allowed people to regain autonomy while reducing fear, anxiety, and uncertainty. The findings from this study suggest that although the Haven still served as a form of temporary shelter, the space enabled participants to experience some of the benefits typically associated with permanent housing, such as rest and stability ( Robinson et al., 2022 ). These benefits had tangible benefits for their health, such as initiating and maintaining healthcare. Previous research identified the need for multidisciplinary collaboration in which services communicate effectively with each other for people experiencing homelessness ( Greene, 2021 ). Our results align with prior research and provide an exemplar that could address these challenges. Optimal collaboration between participants and Haven staff occurred when care was personalized to individual needs and when trusting relationships could be built. On-site integration of case management and nursing care created an environment where people experiencing co-occurring chronic health issues and homelessness could stabilize. Care coordination initiatives that link patients to health and social services are essential to addressing socioeconomic determinants of health outcomes, service utilization, and cost ( Albertson et al., 2022 ; Towe et al., 2020 ). Results from our study are similar to that of a recent systematic review that studied linkage across settings for people experiencing homelessness where three elements of successful care coordination mimicked the model at the Haven (i.e., systematic assessment of patient needs, in-person communication, and standard protocols) ( Albertson et al., 2022 ). While emergency-level operations are not sustainable in the long term, continued coordination is needed between organizations serving persons experiencing homelessness, clinical providers, and health departments to address the complex needs of this population. Model components of the Haven, such as implementing harm reduction strategies and ongoing clinical care management, can be modified and scaled to other existing settings like shelters and residential treatment programs. Co-located, comprehensive service models are somewhat nascent, and evidence on long-term health outcomes is limited ( Melnikow et al., 2020 ). However, there has been promising evidence about elements of such programs, including Housing First and medical respite, that suggest these models may reduce housing instability and inpatient hospital usage ( Biederman et al., 2019 ; Doran et al., 2013 ; Kertesz et al., 2009 ; Shinn & Khadduri, 2020 ). Integrated models require buy-in from health systems and providers to minimize existing barriers and promote specialized care. Increased coordination among stakeholders should address known challenges that impede the continuity of care through building sufficient service capacity, establishing non-congregate spaces to facilitate recovery, and increasing communication across sectors. Promising approaches exist, as those described above, in healthcare, substance use treatment, and housing sectors, but expanded funding streams and implementation that guarantee the sustainability of these efforts are greatly needed. Findings from these interviews also highlighted the potential benefits of an ICS model and the perceived negative impact of the Haven’s closure. The barriers to health experienced by the participants combined with severely limited housing options highlight an urgent need for dedicated programming designed to take care of people experiencing both complex medical needs and housing instability. Expanding programs like medical respite and permanent supportive housing have been shown to be successful in achieving those goals ( Canham et al., 2021 ; Kinczewski et al., 2021 ). In addition, this study highlights the gap in prevention services for those at risk for homelessness, as a onetime response to an ongoing problem, is not sufficient. Further development and expansion of this model will only be possible with sustainable funding sources and will take partnership of housing services and health agencies. Future research is needed to complement the qualitative results of this study and should quantify health outcomes to demonstrate the potential public health impact of this model. The participants in this study were recruited by the staff or self-selected, which may lead to bias toward residents who were more positive about their experience. We also did not yet quantify the specific services received by each participant, and it is likely that the participants varied in their experience of services. The perspectives of administrators and staff would be useful to add to the narrative of residents. This study was conducted a single site in an urban city, and thus generalizability of these findings may be limited with respect to other settings and localities.

Many cities nationwide face a constant and growing crisis of housing instability resulting in families and individuals experiencing homelessness. While challenging to quantify, a single-point-in-time count estimated that there were 580,000 people experiencing homelessness on a single night in January 2020 in the United States ( National Alliance to End Homelessness, 2021 ). It is likely that the incidence of homelessness, and certainly housing insecurity, has further increased due to the societal and economic ramifications of COVID-19 ( Pawson et al., 2021 ; Tsai et al., 2022 ). Homelessness has well-documented negative impacts on health and is associated with high levels of chronic and acute disease, mortality, and cyclical acute care hospital utilization ( Gryczynski et al., 2020 ; Morrison, 2009 ). Integrating healthcare delivery at shelters has been shown to improve both access to healthcare and clinical outcomes for people experiencing homelessness through directly addressing physical and mental health needs with on-site medical assistance and coordinated referral to outside services (Clark et al., 2021; Cornes et al., 2018 ). People with integrated care and permanent housing use emergency departments less often and experience shorter hospital stays ( Sadowski et al., 2009 ; Salhi et al., 2018 ). However, studies have found mixed results regarding the effect of temporary shelters on attaining permanent housing and physical or mental well-being ( Benston, 2015 ). Temporary shelters are often complicated by staff turnover, low resources, and overcrowding, which leads to the spread of infectious diseases ( Moffa et al., 2019 ). Shelters often have rules, such as enforced entry and exit times that may be incompatible with one’s personal daily routines (e.g., work schedule) or need for rest ( Jang et al., 2021 ). Concerns about the safety of one’s personal belongings are also common, as well as concerns about safety from disease and other harm ( Murphy & Eghaneyan, 2018 ; Stewart & Townley, 2020 ). Therefore, recovery after discharge from a hospital stay can be challenging in these environments ( Canham et al., 2019 ). The COVID-19 pandemic necessitated the creative reimagining of shelter services to protect people experiencing homelessness. The pandemic allowed exploring new service delivery models, such as providing integrated care of health needs within a temporary shelter ( Jang et al., 2021 ; Oudshoorn et al., 2021 ; Rosecrans, et al., 2022a ).