Community-level perceptions of housing services for people with opioid use disorder: a qualitative analysis of community stakeholders in the HEALing communities study

Abstract

Background

Housing insecurity, or the inability to afford or maintain residence in a safe and quality place, is a barrier to accessing and staying in substance use treatment and/or recovery support services. There is a gap in knowledge in how housing insecurity influences substance use disorder (SUD) prevalence, and the role housing security plays in SUD treatment initiation and retention. The HEALing Communities Study (HCS) is a large multisite trial aiming to reduce overdose risk through the implementation of evidence-based interventions delivered by community coalitions to reduce overdose deaths among 67 counties in 4 states. This study explored diverse perspectives of HCS community coalition members on housing systems for populations with SUD to better understand how housing insecurity attenuates SUD prevalence and initiation or retention in various drug treatment services.

Methods

The aim of this study was to explore various mechanisms at multiple levels through which housing insecurity adversely affects individuals with SUD. Semi-structured key informant interviews (n = 85) were conduct with HCS community coalition members in New York State from 2019 to 2020. The transcripts were coded using the PRISM/RE-AIM framework and further subcoded with the following five themes: 1) cost (housing affordability); 2) conditions (housing quality); 3) consistency (residential stability); 4) context (neighborhood opportunity); and 5) considerations (unique aspects affecting substance use-related populations).

Results

Findings include housing insecurity driven by cycles of poverty or poor economic reality in NY counties, substandard or poor-quality housing, experiences of stigma or social isolation within the home, a lack of transitional housing for individuals exiting the criminal legal system, housing insecurity associated with a return to use or overdose, neighborhood challenges such as accompanying transportation or food insecurity, or environmental triggers to return to use, and challenges associated with housing insecurity for SUD treatment, such as missed appointments.

Conclusions

The results suggest that structural housing interventions are needed to improve the health and quality of life of individuals with SUD directly and that they may also strengthen the effectiveness and sustainability of other interventions with this population. Future research can explore the community co-design of innovative solutions across the continuum of housing security.

ClinicalTrials.gov identifier

NCT04111939

Introduction

A key affected population among those affected by SUD are those that experience housing insecurity and its most extreme form, homelessness [1, 2]. Housing insecurity, which is defined as the inability to maintain stable and high-quality residence, can include couch-surfing, residing in temporary or transitional housing, squatting in abandoned buildings or homes, or sleeping in one’s car, parks, or other public places [1]. Although prevalence rates vary, it is estimated that about one-third of people with SUD are affected by housing insecurity [3], and 12–13% experience homelessness [4, 5]. Housing insecurity and diagnoses of SUD are often comorbid and bidirectional, with both SUD leading to housing insecurity and housing insecurity leading to the initiation of or increased substance use, often as a coping mechanism [6]. Substantial research has also documented the deleterious impacts of housing insecurity, such as increased mortality [7], and morbidity due to complex health conditions such as diabetes [8], heart disease [9], mental illness [10], and asthma [11]. In addition, the intertwining of housing insecurity, SUD, and mental illness is frequently referred to as a syndemic that is shaped by larger structural factors attributed to extreme poverty, gentrification, residential segregation, discrimination, and other social and political factors related to deprivation [12, 13]. As a result, it is challenging for providers working in SUD treatment to turn a blind eye to housing, especially as housing insecurity is increasing in severity and how it is manifested [14].

The impact of housing insecurity and homelessness on people who use drugs (PWUD) is urgent and severe. In the United States, overdose deaths increasingly occur at home [15], but largely in “marginal housing environments” [16] among individuals experiencing housing insecurity, such as single-room occupancy (SRO) housing (e.g., hotels, motels), supportive housing, transitional housing, and public benefits housing. Overdose death rates among SRO residents from 2010 to 2017 was 20 times higher than that among non-SRO residents [17]. Individuals who are housing insecure have significantly higher rates of overdose than populations who are housing secure (15.8% vs. 3.7%, respectively) [18] and are at 1.5% greater risk of overdose and 8.9% greater risk of hospital admission or emergency department visits related to opioids [19]. People experiencing homelessness are twice as likely to require hospital care [20], with 10 times higher rates of substance use-related hospitalizations and 2 times higher rates of emergency department visits than individuals who are not experiencing homelessness [21].

While addressing housing insecurity from a human rights perspective is important, improving the effectiveness of clinical interventions and the efficiency of healthcare and social service systems is also essential [22], and a way to center housing as a shared goal to foster interdisciplinary collaboration across sectors. Studies have also revealed that homelessness reduces the rate of SUD treatment completion [23], including but not limited to a lack of transportation, the need to prioritize unmet basic needs (e.g., shelter, food), and the experience of stigma from providers [24]. People experiencing homelessness are more likely to access inpatient treatment or medically supervised withdrawal than receive medications for SUD [25]; however, literature indicates that those experiencing homelessness and housed individuals have similar outcomes once they access pharmacotherapy [26]. Therefore, given the comparable treatment outcomes when pharmacotherapy is accessed [26], tailored interventions are needed to increase access to and utilization of evidence-based treatments, particularly medications for SUD, among people experiencing homelessness. These interventions should address the structural barriers—including lack of transportation, competing for basic needs, and provider stigma [24]—that impede treatment engagement.

Experiences of homelessness and housing insecurity are complex problems, occurring within the interplay of multiple individual, interpersonal, and socioeconomic factors [22]. This complexity necessitates solutions that are multifaceted with coordination across systems and appropriate resource levels and allocation. Research evidence indicates that solutions using a “bottom-up” approach versus a “top-down” approach, in which service providers, community-level actors, and consumers are meaningfully engaged, are more effective and sustainable [27, 28]. While they are not directly embedded in housing systems, health and social service providers and other community-level actors are often confronted with the need to address the impact of homelessness and housing insecurity among individuals with SUD, and should be included among partners.

To combat the opioid crisis, the National Institutes of Drug Abuse (NIDA) funded the HEALing Communities Study (HCS), a randomized controlled trial (RCT) testing the effectiveness of the Communities That HEAL (CTH) intervention to reduce overdose deaths through community-led implementation of evidence-based programs (EBPs) in 67 communities across 4 states [29]. Implementation was facilitated by identifying and addressing barriers to and facilitators of the implementation of medication for opioid use disorder/overdose education and naloxone distribution (MOUD/OEND). As part of the broader HCS effect, baseline qualitative interviews were conducted with selected HCS coalition members, including but not limited to community champions and representatives from health and social service organizations in an area. This qualitative analysis aims to synthesize the perspectives related to homelessness and housing insecurity and to provide a more in-depth characterization of the patchwork system of housing resources and actors that intertwine with the lives of people with OUD in New York State.

The theoretical framework guiding HCS is the Practical, Robust Implementation and Sustainability Model/Reach, Adoption, Effectiveness, Implementation, Maintenance (PRISM/RE-AIM) framework [30, 31], which posits that key external and internal contextual factors influence the implementation of evidence-based interventions. Housing is consistently discussed as a key external contextual factor; as a result, the aim of this study is to more deeply investigate perceptions of housing among community-level actors, to precisely characterize key factors for populations with SUD. To better understand and characterize housing, we use Swope & Hernandez’s [32] conceptual model for housing as a determinant of health equity, which posits four pillars of structural inequalities that shape the unequal distribution of access to health-promoting housing factors: 1) cost (housing affordability); 2) conditions (housing quality); 3) consistency (residential stability); and 4) context (neighborhood opportunity). The health equity model conceptualizes these pillars as interacting elements contributing to cumulative burdens that “produce and reify health disparities.” We share Swope and Hernández’s view that “housing is a critical pathway through which such disparities develop and is potentially an effective means for ameliorating them through a comprehensive understanding of the complexities of housing” (2019). The health equity framework incorporates policy and community-level factors such as political mobilization, social views, and context-specific policies and practices that shape cities, such as redlining [32]. Thus, this model provides a solid foundation for exploring context-specific community perspectives on housing complexities with a lens toward health equity.

Materials and methods

Study design and setting

The HEALing Communities Study is a multisite, wait-listed, community-level cluster randomized trial that aims to test the effectiveness of a community-level intervention on reducing opioid deaths in communities that have been heavily impacted by the opioid epidemic [33]. The HCS study was carried out in four states from April 2019 to December 2023, with a high burden of opioid use disorder (OUD) and overdose deaths, one of which is New York [33]. The present analysis was restricted to data from NY, where 16 counties were chosen, with variations in terms of urbanicity, racial and ethnic diversity, and other socioeconomic factors, such as extreme poverty and high rates of unemployment. Participating HCS counties in New York State also have significant housing burdens, with the percentage of cost-burdened households (defined as paying more than 30% of their annual income on housing) ranging from 21.7% to 37.5%. In addition to the cost burden, the number of households that report overcrowding or incomplete kitchen or plumbing facilities ranges from 0.9% to 2.3%, indicating high levels of substandard living facilities or outdated housing stock (Table 1).

Table 1.

HCS New York state-specific county descriptive statistics related to housing (N = 16)

Wave	Urban/Rural	% of cost-burdened households1	% of households reporting problems2	Diff	% of severely cost-burdened households1	% of households reporting severe problems2	Diff
2	Urban	27.2%	28.7%	1.5%	13.0%	14.7%	1.7%
1	Rural	24.9%	26.3%	1.4%	10.8%	12.4%	1.6%
2	Rural	25.5%	27.1%	1.6%	12.3%	14.4%	2.1%
1	Rural	26.7%	28.0%	1.3%	13.4%	14.9%	1.5%
2	Rural	23.5%	25.2%	1.7%	10.5%	12.9%	2.4%
1	Urban	25.8%	27.1%	1.3%	13.0%	14.8%	1.8%
2	Rural	22.3%	24.2%	1.9%	9.9%	12.4%	2.5%
1	Rural	29.2%	30.5%	1.3%	14.4%	16.0%	1.6%
1	Rural	21.7%	22.6%	0.9%	10.2%	11.5%	1.3%
2	Urban	28.6%	29.6%	1.0%	14.0%	15.5%	1.5%
2	Urban	37.5%	39.8%	2.3%	17.6%	20.9%	3.3%
1	Urban	35.6%	36.8%	1.2%	15.2%	16.7%	1.5%
1	Urban	36.4%	38.1%	1.7%	17.5%	19.8%	2.3%
2	Rural	29.5%	30.8%	1.3%	13.5%	15.1%	1.6%
1	Urban	34.2%	35.6%	1.4%	16.7%	18.6%	1.9%
2	Urban	25.4%	27.1%	1.7%	10.2%	12.4%	2.2%

¹ Cost burden is measured as the ratio of housing costs to household income > 30% (severe > 50%) (Source: US Department of Housing and Urban Development, Comprehensive Housing Affordability Strategy data, based on 2016–2020 American Community Survey estimates)

² Housing problems are defined as one or more of the following: incomplete kitchen facilities, incomplete plumbing facilities, > 1 person per room, and cost burden > 30% (severe > 50%) (Source: US Department of Housing and Urban Development, Comprehensive Housing Affordability Strategy data, based on 2016–2020 American Community Survey estimates)

This analysis draws from 84 in-depth interviews completed at baseline as part of the HEALing Communities Study (HCS) across 16 counties in New York State. Details about the semi-structured qualitative interview guide, informed by constructs from the PRISM/RE-AIM framework, are published elsewhere [34]. The protocol and related procedures were approved by a central Institutional Review Board (IRB; Advarra Inc.).

Participant recruitment

Individuals from all communities included in the HCS were eligible for interview participation included: (1) coalition members—individuals formally participating in community coalitions addressing the opioid epidemic; (2) community champions—informal leaders advocating for opioid-related initiatives; and (3) key stakeholders—individuals in decision-making positions relevant to the opioid response. Participants were purposefully sampled to obtain broad community perspectives. Participants represented diverse organizational types, including behavioral health, public health, emergency response, criminal justice, education, and community groups. Individuals were recruited to participate in an approximately 60-minute semi-structured interview with an initial email invitation describing the study, followed by reminders via email and telephone [30, 31, 35].

Data collection

The data were collected between November 2019 and January 2020 by a team of research staff trained by senior qualitative researchers. Interviews were conducted either in-person or remotely via telephone or Zoom video conference. Verbal consent was obtained during remote interviews and documented in writing. All interview audio recordings were cleaned, anonymized and de-identified, transcribed verbatim by a professional transcription company, and reviewed for accuracy by a member of the research team. The participants received a $50 incentive after completing the interview unless they declined or were not able to accept compensation due to their position or job. Additional information about the qualitative data collection process and other study methods are published elsewhere [35].

Analytical approach

Data analysis consisted of a two-cycle coding process to arrive at consensus. A codebook with a priori codes based on the domains of the PRISM/RE-AIM framework [30, 31, 35] was created, containing four parent codes: 1) external context, 2) internal context, 3) intervention and implementation, and 4) outcomes. Research staff engaged in preliminary coding via NVivo 12 (QSR International, Melbourne, Australia) to establish cross-site consensus with the other three states (KY, OH, and MA) and met weekly to ensure coding consistency, which was reviewed by senior qualitative researchers, after which transcripts were divided and independently coded [36]. A smaller group of research staff then further sub-coded the “community risks” child code of the “external context” parent code of the baseline interviews via NVivo 14 (QSR International, Melbourne, Australia). A sub-codebook was developed with a priori codes about homelessness, housing services, poverty, other social determinants of health, and a section for emergent inductive codes from the data. Transcripts were tripled coded by the smaller group of research staff, and discrepancies were identified and resolved by group consensus. Sub-codes were combined into salient themes and subthemes on the basis of the four pillars of the housing and health equity framework [32] and included in analytical matrices along with interview excerpts illustrating them. The results were iteratively reviewed, interpreted, and vetted by several senior qualitative researchers and HCS community coalition members. Demographic survey data were analyzed via descriptive statistics and the Statistical Package for the Social Sciences (SPSS) software, version 25.0 (I BM, Armonk, NY).

Results

Sample and characteristics

Key informants were approximately half female (n = 44; 56.4%) and half male (n = 34; 43.6%) (see Table 2). Most were middle-aged, with over two-thirds of the sample being 35–49 years of age (n = 28; 33.3%) or 50–64 years of age (n = 29; 34.5%), followed by those 65–74 years of age (n = 17; 20.2%) and 18–34 years of age (n = 10; 11.9%). More than half of the key informants (n = 45; 57.5%) had completed a Master’s degree, followed by a Bachelor’s degree (n = 16; 20.5%). The majority of key informants identified as White (n = 76; 90.5%) and not Hispanic or Latino (n = 76; 90.5%). There were a broad range of community perspectives, including those working in harm reduction (n = 24; 28.9%), government or policy-related positions (n = 21; 25.3%), criminal justice (n = 11; 13.3%), and healthcare (n = 10; 12.0%).

Table 2.

Demographics of key informants in HCS community coalitions (N = 85)

	n (%)
Age (years)1
18–34 years	10 (11.9%)
35–49 years	28 (33.3%)
50–64 years	29 (34.5%)
65–74 years	17 (20.2%)
Race2
Asian	1 (1.2%)
White	76 (90.5%)
Native Hawaiian or Other Pacific Islander	6 (7.1%)
Hispanic or Latino/a
Yes	76 (90.5%)
No	8 (9.5%)
Gender3,5
Male	34 (43.6%)
Female	44 (56.4%)
Education level4,5
High school degree or equivalent (e.g. GED)	2 (2.6%)
Some college, no degree	1 (1.3%)
Associate degree (e.g. AA, AS)	2 (2.6%)
Bachelor’s degree (e.g. BA, BS)	16 (20.5%)
Master’s degree (e.g. MA, MS, MEd)	45 (57.7%)
Professional degree (e.g. MD, DDS, DVM)	2 (2.6%)
Doctorate (e.g. PhD, EdD)	10 (12.8%)
Organization6
Behavioral health	3 (3.6%)
Community group	3 (3.6%)
Criminal justice	11 (13.3%)
Education	2 (2.4%)
Government/policy	21 (25.3%)
Harm reduction	24 (28.9%)
Healthcare	10 (12.0%)
Other	9 (10.8%)

¹ Categories not shown had zero frequency (75 and older)

² Categories not shown had zero frequency (African American/Black; American Indian/Alaskan Native)

³ Categories not shown had zero frequency (trans male/trans man; trans female/trans woman; genderqueer/gender nonconforming)

⁴ Categories not shown had zero frequency (less than high school)

⁵ Percentages were calculated excluding missing data (n = 6), valid n = 78

⁶ Percentages were calculated excluding missing data (n = 6), valid n = 78

Themes

Figure 1 presents a summary of the main themes emerging from the qualitative analyses by area of analysis corresponding to the housing and health equity framework [32].

Fig. 1.

Summary of the perspectives of community coalition members about homelessness and housing insecurity

First pillar: Cost (housing affordability)

Housing affordability is whether individuals or households can pay the cost of housing without burden [32]. Community coalition members discussed the lack of availability of housing options across the spectrum, ranging from services for people experiencing homelessness to affordable housing, as well as unique types of housing services such as transitional housing for individuals who were recently incarcerated, and homeless services.

“And so, that’s one of the challenges that we work around. Housing including homeless services and temporary transitional housing is in short supply. But also just safe affordable housing is a real need in our community”. (HCS #734)

Study participants also discussed the broader economic reality underpinning housing affordability in their respective counties, and how the unavailability of jobs that pay a living wage translates to limited options for housing, or no alternative other than public housing.

“Because of their transportation problems and because of the economic reality of [where they live is] very low paying jobs, jobs that in most places will not get them the ability to afford better than social agency housing”. (HCS #050)

In addition to employment, community coalition members discussed the interdependence of housing with other social determinants, such as food security, and how they were interconnected with extreme poverty.

“There’s terrible cycles of poverty. Terrible – and again, you know, it’s all connected, in my opinion … I don’t have housing, I don’t have adequate food, a good paying job … it’s all related, you know”. (HCS #927)

Second pillar: Conditions (housing quality)

Housing quality is the adequacy of the physical and environmental conditions of the building and housing unit [32]. Community coalition members discussed the inadequacy of public housing available, and the resulting challenges for individuals living in extreme poverty.

“They have housing challenges … if you’ve ever been in Section 8 housing, there is such a thing as real Section 8 housing where somebody actually inspects and they do follow. Then there’s just housing that Section 8 says, ‘Oh, we’ll take this, this is the best we got’.” (HCS #050)

In addition to the suboptimal physical conditions of housing, community coalition members discussed the social inadequacies of available housing for individuals with SUD, such as experiences of stigma, lack of social support, and isolation.

“When you send them home, because you can’t get them an apartment, not all families are particularly understanding, not all spouses are particularly understanding or significant, not all families are safe to go back to … because it doesn’t mean they’re not using something else. In rural counties that’s really tough because isolation is really easy”. (HCS #050)

Third pillar: Consistency (residential stability)

Residential stability is the individual or household’s ability to remain in their home for as long as they desire [32]. Key informants discussed the importance of safe and secure housing to reduce the risk of overdose death; however, housing providers struggle financially to stay open.

“The supportive services out in the community really struggle to maintain adequate funding for people who are in recovery but need a temporary period of community residence. They are also some of the most challenging people. Within the last year and three months, we’ve had three deaths of opioid dependent people … these are people that were stable in recovery doing well, training and services, who, once they relapsed, they died … which is a common pattern of people who come out of, whether it’s penal institutions or community residences with long histories of abstinence, but find themselves relapsing. And, they – they overdose. And so, addressing the needs of that population is a strong commitment that [de-identified temporary housing organization] has, but is struggling to maintain financial, uh, solubility … to be able to do so”. (HCS #785)

Community coalition members also discussed the importance of stable transitional housing for individuals exiting the criminal legal system, which is also essential for continuity of care.

“It’s trying to implement things to ease that transition for success from jail. So we accept public assistance applications 30 days prior to the planned discharge from jail. So, I look at jail as a discharge like you would in health care, you know? [How] do we plan for you to be successful instead of, ‘Oh, here comes another criminal,’ right? How do we minimize homelessness? How do we minimize shelter stay? How do we engage those that were getting treatment in jail so that there isn’t a gap?” (HCS #067)

Fourth pillar: Context (neighborhood opportunity)

Neighborhood opportunity is the presence of health-promoting or health-limiting infrastructure and resources in the surrounding area [32]. Community coalition members describe neighborhoods in negative contexts, alluding to a multitude of environmental factors (e.g., crime, lack of infrastructure) that affect individuals with SUD.

“I think there maybe one halfway house in [a different county]. And that’s in a [city] that’s not really an area that people wanna go to ‘cause it’s pretty bad”. (HCS #945)

Key informants also discuss community-level stigma, one manifestation of which is not in my backyard, or NIMBY, mentality, which hinders implementation of innovative housing solutions.

“The not in my backyard [NIMBY] type of mentality … But it’s such a small county, there is no way that’s not somebody’s backyard. [People] have been trying to put a homeless shelter in this county for a long time. No one will let them do it anywhere. You got people who have millions of dollars like available to build, you know, subsidized housing for lower income that would also house people with mental health challenges or people in recovery. No one will let them do it. No one will let them build it. Cannot find a site where a town [that will] allow them to do it. So, it’s not necessarily [that] they’re not acknowledging the problem of homelessness or mental health, but when it comes down to action, it doesn’t seem like people want those types of things that are interesting”. (HCS #326)

Community coalition members discuss the lack of public transportation as a barrier in neighborhoods where families who are in recovery reside.

“They really do not have places for women with lots of children, maybe one, when you get to two or more becomes difficult, and if you get beyond three, it can become very difficult to find housing, inexpensive housing because they have to get back on their feet … and they don’t have licenses and there’s really very little public transportation [where they live]”. (HCS #050)

In addition to transportation, key informants discussed other social determinants of health, such as food insecurity, and the interdependence of these factors attributed to living in extreme poverty.

“We’re addressing the collective … our conversation today is geared towards one head of the dragon. But this is like an eight-headed dragon. So if we don’t address the homelessness, if we don’t address the hunger, if we don’t address the fear, the question that we always ask people is not, ‘Why do you get high?’ The question is, ‘Why wouldn’t you get high?’ [There’s] a lot of poverty in the county”. (HCS #556)

Fifth pillar: Considerations (unique aspects among substance Use-involved populations)

Themes were also identified that were unique to people who use drugs. For example, community coalition members perceived a cyclical nature of homelessness and substance use issues.

“What I hear from everyone is, ‘We don’t know how to break the cycle for people’.” (HCS #723)

Key informants use “homelessness” as a descriptor of hard-to-reach populations with multiple co-occurring conditions that are affected by SUD and end up utilizing emergency health services.

“The consortium is our frontline workers and a frontline supervisor talking about these incredibly hard-to-serve adults [that] are failing in the community who we can’t engage in either mental health or substance abuse. And they’re high utilizers of either health care or EMS services or home – they’re constantly homeless”. (HCS #067)

The participants also discussed the lack of availability of unique housing services specific to SUD treatment. Overcrowding and the need to travel far distances to access appropriate housing options were noted.

“The one thing that I would say that I definitely think is missing in our community now I think of it is like after care. There’s no halfway houses, there’s no sober living environment. There’s no place for people to go after they got their treatment, you know … [what is available is] super far away. Like I know, like my local town, like let’s say, someone lived here and they went to treatment. And they, they don’t wanna come back here. I think the closest place for them to go if they can’t find a bed because the places are so overcrowded, um, is in like a different county”. (HCS #945)

Housing challenges also comprised treatment attendance and subsequent health outcomes. Missed appointments were often the result of being housed away from services with limited transportation options.

“‘Cause I know a lot of them, the struggle was, you know, with housing, with transportation and with all these appointments scattered throughout the day, trying to make all these different appointments, that’s where they really kind of fell off …”(HCS #751)

In addition, there is provider-level stigma about individual-level decision-making versus the structural nature of housing disparities driving these issues.

“They’re homeless and they don’t want to try to work on any issues … [They] may comply for a week, and then they’re back putting themselves in jeopardy, frankly”. (HCS #723)

Lastly, community coalition members also discussed stigma as a barrier in their work advocating for substance use-involved populations, especially public dollars spent on housing.

“I’ve seen a good number … at DSS [Department of Social Services] in halfway and three-quarter houses. They’re going to treatment – they’re going to be in-patient, coming out and needing half or three-quarter housing … to be successful … Our elected officials and our legislature [will] have a conversation with our community services director and all are gung ho on treatment until they realize maybe DSS is paying for the treatment or the place for them to live while they do intensive treatment. So, that seems to be like, ‘Oh, you’re spending money to do it?’ Like … [laughing] and it’s suddenly ‘those people’ instead of, you know, someone needing assistance”. (HCS #067)

Discussion

This qualitative analysis unpacks mechanisms at various levels through which housing insecurity adversely affects individuals with OUD. Themes related to poverty and poor economic conditions, substandard or poor-quality housing, instability related to dwelling, neighborhood-level stigma and environmental factors and triggers were discussed in how they coalesce to affect substance use prevalence and serve as barriers to SUD treatment initiation and retention. Community coalition members largely perceived the importance and urgency of improving housing systems for people with SUD. The findings from this study underscore the need for and importance of efficacious, cost-effective housing interventions to reduce overdose risk and increase access to and retention in treatment and utilization of harm reduction services. The results suggest that structural housing interventions not only improve the health and quality of life of individuals with OUD directly but also increase the effectiveness and sustainability of other interventions targeted at this population [37]. The interviews with community coalition members demonstrated their knowledge of the cyclical nature of housing and substance use risk and the need for structural interventions to interrupt these patterns. These findings are consistent with previous studies on housing and substance use [38, 39], homelessness and overdose risk [40], the importance of housing security for substance use treatment support services [41], and the call for structural interventions to improve housing among this population [12, 42].

Access to affordable, safe, and secure housing is an essential element of well-being, particularly for people dealing with chronic health conditions such as OUD and other SUDs. Housing can be considered an essential prerequisite to health but not a feature that is sufficient to guarantee health on its own [32]. Housing is often conceptualized as the most important determinant of health [43] and the anchor leading a livable life [42, 44]. This analysis builds upon the work of other researchers who have conceptualized relationships between social determinants of health and structural features of a community (e.g., policy, available services, economic landscape), as varied and interrelated. Other frameworks applied to qualitative inquiry connecting housing and health have conceptualized the importance of housing related to social cohesion [45] and less tangible aspects of housing (e.g., tenancy experience, neighborhood and support networks) to health outcomes [46]. Context is also essential when incorporating considerations for substance use-involved populations into conceptualizations of housing. For example, using substances by oneself increases the risk of overdose [17]. Therefore, access to single-room housing may not be effective in creating safety for people who use drugs, particularly in the absence of embedded substance use treatment and harm reduction services [47]. To understand these complex contextual factors, the perspectives of substance use-involved populations and professionals working directly with them are essential [48].

This study illuminates the role of stigma toward those experiencing homelessness and substance use, a finding that is supported by the literature [49, 50]. A recent study conducted in Los Angeles revealed that people experiencing homelessness report levels of weekly discrimination (54%) and monthly physical and sexual violence (16%) that most populations experience over the course of their lifetime [51]. Findings from this analysis also document the stigmatizing and paternalistic way in which providers discuss people experiencing homelessness, an important consideration, as provider-level stigma could hamper the quality of care delivered [52] and community advocacy efforts [53]. Furthermore, stigma can act as a barrier to care. One study among individuals experiencing homelessness with SUD revealed that experiencing stigma from care providers toward people who use drugs was a barrier to MOUD [54]. In addition to housing, addressing stigma from healthcare professionals to improve treatment uptake and retention may be important.

The study findings indicate a significant complexity between transitioning out of the criminal legal system and housing insecurity among PWUD. Multiple community coalition members discuss re-entry and transitional housing services as a significant component of the housing service system for PWUDs, an area that has many gaps where individuals fall through systems of care. People with criminal legal system involvement face additional challenges where substance use can lead to incarceration and where prior incarceration can impede stable housing [55]. The criminalization of substances has done little to decrease their use and has resulted in family separation, community division, and disproportionate incarceration of minoritized populations [56]. Literature also documents the spillover effect of incarceration, with the partners and family members of incarcerated individuals also experiencing housing insecurity [57]. Housing and urban development in the United States are largely driven by racial capitalism [58], resulting in homelessness service systems that professionalize this crisis instead of being incentivized to end it [22]. The application of abolitionist principles to these systems requires pro-health and pro-housing logic to achieve the goals of health equity and disparity elimination [59, 60].

Coalition members discussed the importance of housing conditions, capturing not only physical inadequacies (e.g., safety from weather or natural elements) but also social inadequacies, such as a lack of social support or safety from interpersonal or community violence. Further research qualifying or quantifying safety, especially from different intersectional perspectives (e.g., across considerations of gender, race, ethnicity, and family status), is needed to design housing interventions that are adequate for people experiencing homelessness. Research has revealed that people experiencing homelessness has a different calculus of safety and risk [61], and the “security” of housing provided by service providers, even those with a Housing First Model, is documented to fulfill other psychosocial needs [62]. Research documents that people experiencing homelessness perceived congregate homeless service settings as restrictive, punitive, and undermining [63], resulting in people experiencing homelessness frequently and unsustainably shuffling between different homeless service providers and other temporary living institutions [64].

Limitations

This study has an important limitation. These interviews are only from opioid overdose prevention coalition members at the community-level, and perhaps diverge from perceptions of housing systems among people with OUD [65]. Although this topic emerged as an important theme, the interviews themselves did not explicitly focus on housing; therefore, we may have captured only surface-level perceptions instead of depth in relation to the issue of housing systems. Stigma operates more so in open-ended interviews than in more focused data collection, with interviewees not always distinguishing themselves from their personal views and professional expertise (e.g., internalized biases that seep into their work) [66]. In addition, the sample of coalition members is disproportionately white, and the lack of inclusion of members identifying as part of racial minoritized groups may result in the exclusion of important perspectives on racial inequities in housing security. The overall HCS study did not interview people with OUD, and it is a limitation to not be able to triangulate these community-level findings with individual-level experiences.

While the goal of qualitative research is not to achieve generalizability, it is important to keep in mind that the perspectives of community coalition members in this study may not reflect what other staff within and outside of these organizations think regarding housing systems for populations with SUD. Similarly, the findings may not be generalizable to other U.S. locales with different demographics, levels of urbanicity, or local and state policies and politics. The participants’ responses could also have been biased due to social desirability or out of caution; however, study procedures to guarantee the anonymity and privacy of these interviews, especially the assurance of confidentiality during the informed consent process, mitigate the concern for this bias. Our findings reflect the importance of housing systems tailored to considerations of substance use in rural areas, but there may be other important variables that render individuals with SUD more likely to struggle to find and access services than others. Future research can unpack and explore the interdependence of these factors to improve service delivery for this population.

Conclusion

This qualitative study explores perceptions of the severity and urgency of housing for individuals with substance use issues, as perceived by community coalition members from a variety of sectors participating in a large federally funded multi-site trial. The findings are reflective of the complexity of housing systems and the needs of housing across the continuum, ranging from temporary and transitional, permanent supportive housing, to independently and stably housed. There are very few protections in political or economic structures that guard against homelessness in the U.S., which, coupled with the pervasive view of homelessness as an individual problem resulting from bad luck or poor choices, results in significant stigma, which acts as a barrier to community-level housing solutions [43]. Despite this, interventions that focus on housing have been found to be effective in both improving SUD-related outcomes and housing stability. Effective interventions for this population are specifically targeted at low barriers, and interventions focused on housing have also been found to be beneficial [25]. Multiple housing approaches exist: the Housing First model, which focuses on immediate access to housing without the requirement of sobriety or participation in supportive services, and the recovery housing model which emphasizes abstinence and safe housing for individuals experiencing homelessness and SUD [67]. Housing Choice, a program piloted in Portland, OR, is a combined approach that incorporates elements of both recovery housing and Housing First to allow individuals to select housing based on their personal needs. Individuals can choose from a range of housing options, including transitional housing, PSH, family housing, Housing First, and recovery housing. An RCT is in progress [67] examining the effectiveness of various types of housing interventions. Studies have shown that pairing assisted housing with medication increases treatment retention [68] and that engaging in housing programs can reduce opiate or stimulant use [69]. Interestingly, one study revealed that states with fewer low-income housing tax credit units had more opioid overdose emergency department visits than did states with more units. This finding indicates that greater provision of affordable housing may prevent opioid overdoses [70].

Housing interventions that do not include the involvement of people experiencing homelessness or people with lived experience in program decision-making have lower fidelity, inhibiting proper implementation [71]. Future research can explore the community co-design of innovative solutions across the continuum of housing stability. There is a great need for capacity building among health and social service providers. Housing is a basic human right and deserves to be a policy priority at the local, state, and federal levels for populations with SUD and other co-occurring chronic conditions. Future research can further explore the causal mechanisms of how housing plays a role in the prevention and treatment of SUD, along with opportunities for intervention to improve the implementation of programs across the continuum of care.

Abbreviations

SUD

Substance use disorder

HCS

HEALing Communities Study

PWUD

People who use drugs

SRO

Single-room occupancy

NIDA

National Institute of Drug Abuse

RCT

Randomized controlled trial

CTH

Communities that HEAL

EBP

Evidence-based program

MOUD

Medication for opioid use disorder

OEND

Overdose education and naloxone distribution

PRISM/RE-AIM

Practical, Robust Implementation and Sustainability Model/Reach, Adoption, Effectiveness, Implementation, Maintenance

Author contributions

N.E. and L.G. acquired the funding for the overall HCS project. N.D. and S.B. conceptualized the study. N.D., S.B., A.D., D.G-E., and L.G., designed the methodology. N.D., S.B., and M.R. conducted the data analysis and wrote the main text of the paper. N.D. created the table, and N.D. and D.H. created the visualization. All authors reviewed and edited the manuscript.

Funding

This research was supported by the National Institutes of Health and the Substance Abuse and Mental Health Services Administration through the NIH HEAL (Helping to End Addiction Long-term®) Initiative under award numbers UM1DA049415 (ClinicalTrials.gov Identifier: NCT04111939). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, the Substance Abuse and Mental Health Services Administration or the NIH HEAL Initiative®. One of the authors, Nishita Dsouza, is also supported by the National Institute on Drug Abuse of the National Institutes of Health under Award Number T32DA037801.

Data availability

Data and documentation may be available upon request from the NY state site representative (New York: jld2023@columbia.edu).

Declarations

Ethics approval and consent to participate

This study protocol (Pro00038088) was approved by Advarra Inc., the HEALing Communities Study Single Institutional Review Board. In accordance with the Declaration of Helsinki, every subject in the study provided their consent to participate.

Consent to publish

Not applicable.

Competing interests

The authors declare no competing interests.