Exploring Harm Reduction in Supportive Housing for Formerly Homeless Older Adults.

Background: Exclusionary care policy contributes to the growing number of older adults experiencing homelessness and complex health challenges including substance misuse. The aim of this study was to examine how harm reduction policy and practices are experienced and enacted for older adults with homeless histories and care staff in congregate supportive housing.
Methods: Drawing on harm reduction (HR) principles, Rhodes' risk environment framework, and 15 semi-structured interviews (six residents, nine staff) at a 70-bed supportive housing facility in Western Canada, this qualitative constructivist grounded theory study aimed to determine: How is harm reduction experienced and enacted from the perspectives of older adults and their care staff?
Results: HR policy and practices helped residents to feel respected and a sense of belonging, due largely to staff's understanding of structural vulnerability related to homelessness and their efforts to earn and maintain residents' trust. Physical and program structures in the facility combined with the social environment to mitigate harms due to substance- and nonsubstance-related risk behaviours.
Conclusion: HR policy and practices in supportive living empower care providers and older adults to work together to improve housing and health stability. Wider adoption of HR approaches is needed to meet the needs of a growing number of older people experiencing homelessness and substance use challenges.

INTRODUCTION

Single adults over the age of 50 account for the fastest-growing demographic in North American homeless shelters.( Pathways to homelessness in later life are diverse and multifaceted and include individual, relational, and structural drivers: poverty, unemployment, and lack of affordable housing; social isolation and discrimination; and complex chronic physical and mental health problems including substance misuse.( Homelessness is associated with premature onset of some geriatric conditions such as functional, cognitive, and sensory impairment, as well as frailty, depression, and urinary incontinence, occurring at rates higher than seen in housed adults 20 years older.( The age of 50 is widely used to define “old” for street-affected people who experience higher rates of acute care use and premature mortality compared to their non-homeless counterparts.(

Most emergency homeless shelter and aged care systems contribute to ‘structural vulnerability’ (Table 1), as they do not reflect the needs of older people experiencing homelessness.( Shelter policies (prohibiting walkers and home oxygen), operational rules (no day access, maximum stay times), and physical features (bunk beds, stairs) are ill-suited to older adults who are also vulnerable to victimization by younger shelter users.( Minimum age requirements (usually 65) and “zero tolerance” policies on substance use exclude many homeless older adults from aged care settings, in which unsanctioned use of substances (including tobacco), can be grounds for eviction.(

TABLE 1

Definitions of key concepts that informed data gathering and analysis

Structural Vulnerability	Conceptualization of vulnerability, or health inequalities, in terms of the institutions (e.g., health, social, economic) and social conditions (e.g., classism, ageism, racism, sexism, heteronormativity, colonialism) that determine distribution of health-related resources.(10)
Harm Reduction	is grounded in principles of justice, human rights, autonomy, choice, incrementalism and pragmatism; includes policy and practice which emphasize reducing negative consequences (including health, social, and economic) of substance use without requiring reduced substance consumption.(17–18)
Rhodes’ Risk Environment Framework	The ‘risk environment’ is a product of the complex interplay between social, physical, economic, and policy factors in which risks of harms from substance use are exacerbated. Rhodes (2009) states: “a ‘risk environment framework’ envisages drug harms as a product of the social situations and environments in which individuals participate. It shifts the responsibility for drug harm, and the focus of harm reducing actions, from individuals alone to include the social and political institutions which have a role in harm production” (p193).(19)For example, a large lodge with a single outdoor smoking area can be a ‘risk environment’ for an older person who is tobacco-dependent and has significant mobility challenges. Inability to ambulate to the designated area to smoke multiple times a day, may contribute to unsanctioned indoor use and serious consequences (including eviction in some settings). Interacting elements of the policy and physical environments have combined here to increase the harms associated with smoking. Provision of sheltered, accessible indoor and outdoor locations to smoke would help to prevent such harms to people who are unable or unprepared to abstain from smoking.

Geriatric medicine and gerontology are calling for wider use of harm reduction (HR) approaches (Table 1), especially in the care of older adults with substance use challenges for whom abstinence-only approaches have failed.( While there is a growing body of literature on younger people’s experiences of HR in housing, there has been little study of older people.( To inform policy and practice, the aim of this qualitative study was to determine how HR policy impacts care delivery in supportive housing targeting older people with experiences of homelessness. Of particular interest was how HR policy is experienced and enacted by older people with homeless histories, as well as staff, at a 70-bed supportive housing facility in Western Canada, given here the pseudonym ‘Harbour House’.

METHODS

Study Design and Setting

This study used constructivist grounded theory, which is well-suited to identifying a range of experiences related to little-known social processes.( Data collection and analysis were informed by sensitizing concepts,( including harm reduction principles (choice, dignity, autonomy)( and ‘Rhodes’ risk environment framework’( (Table 1). Rhodes’ Risk Environment Framework draws attention to how the environment in which a risk behaviour such as alcohol or drug use occurs influences the harm individuals may experience.( This study received University of Calgary Conjoint Health Research Ethics Board approval.

Harbour House is a 70-bed care facility, providing long-term housing, social supports, and health care, applying HR policy to older men and women (aged >55yrs) with experiences of homelessness and complex health challenges, including active substance use. Residents have access to meals, house-keeping, and support workers around-the-clock, and a full-time social worker employed by the non-profit agency which operates the facility. Health services are provided on-site by visiting agencies. Residents can participate in Managed Alcohol and Managed Tobacco Programs. Staff procure, store, and dispense alcohol and cigarettes on behalf of residents voluntarily participating in these programs. Alcohol consumption is permitted in private rooms and residents may have guests in their room while consuming alcohol. Tobacco use is permitted in an indoor smoking area and outdoors. Opioid agonist treatment is also available with on-site dispensing. Since opening in 2005, Harbour House has had a 6–18 month wait list for beds. The length of stay is not limited.

Participant Recruitment

Full-time and visiting staff working at Harbour House for one year or more were invited to participate in the study. In consultation with the Harbour House social worker, all residents who had resided there at least six months and were able to participate physically and cognitively were invited by letter from the researchers to an in-depth interview. Residents were given a $25 honorarium for their time; staff were not compensated.

The research team comprised Lara Nixon, an academic family physician who provided primary care at Harbour House, and Victoria Burns, an academic social worker unknown to staff and residents. Both have experience in qualitative research. Nixon was on sabbatical at the time of the study.

Data Collection and Analysis

Data collection and analysis were iterative. All participants were interviewed face-to-face by the two authors at a time/place most convenient to the participants. The semi-structured interviews lasted 45 minutes to 1.5 hours, were audio-recorded and transcribed. All participants were interviewed about their experiences of HR at Harbour House. This included describing what HR meant to them and their experiences of HR during a typical day (see Appendix A). Both investigators made independent field notes during and after each interview.

With constructivist grounded theory, data collection and analysis are conducted simultaneously through theoretical sampling and constant comparison method of data analysis.( Transcripts were coded independently, using line-by-line coding, constant comparative analysis, and sensitizing concepts, which helped to bring initial descriptive codes up to a higher level of abstraction.( Investigators met regularly to compare coding, discuss emerging ideas, and develop focused codes and analytic categories. This supported comparing perspectives and reflexivity. Early findings were presented to residents and staff at Harbour House in two presentations, as a form of triangulation and member checking.( This iterative process continued until both authors were satisfied with the theoretical depth and richness of the analysis, and theoretical saturation was reached.(

RESULTS

Participants’ Characteristics

Of 16 eligible resident participants, eight expressed interest but two subsequently became ill, resulting in six being interviewed. Residents all had multimorbidity, mild-to-moderate frailty, and varied homeless histories including age at first homelessness. The majority had experienced a hospital stay of greater than six months prior to securing a bed at Harbour House and were referred directly from the hospital. Most used alcohol and tobacco, and half reported illicit substance use. Of the 12 eligible staff, all agreed to be interviewed and nine were purposively sampled to represent different types of providers. Further participant demographics and interview details are described in Tables 2a and 2b.

TABLE 2a

Residents’ demographics and interview duration

Resident Participants (N=6)
Gender	3 women, 3 men
Average age (range)	65.8 (58–78) years
Age at time of entry to Harbour House, average (range)	60.5 (55–70) years
Tenure at Harbour House, average (range)	5.3 (2–9) years
Place of residence immediately prior to Harbour House	Shelter (1)Apartment, evicted (1)Hospital with non-acute stay, ≥ 6 months (4): 6–12 months (3); 12–24 months (1, psychiatric admission)
Age at first homelessness,a in years	<20 (1)20–40 (2)40–60 (2)>60 (1)
Places resided while homeless	Rough sleeping, e.g., on street, encamped, in car (3)Emergency Shelters (5)Temporary accommodation, e.g., motel (2)Precariously housed, risk of imminent eviction (6)Provisionally housed, e.g., in hospital >6 months (4)
Current Substance Use	Tobacco (5); Managed Tobacco Program (0)Alcohol (5); Managed Alcohol Program (1)Opioids (1); Methadone (1)Illicit substances (3)
Self-rated Health Status (Excellent-Good-Fair-Poor)	Good (4) Fair (1) Poor (1)
Multimorbidity (≥ 2 chronic conditions)	Yes (6)
CHSA Frailty Score (1–7),b average (range)	5.8 (5–6)
Self-reported mental health challenge	Yes (1), Bipolar disorder
Ethnicity; preferred language	Caucasian (5), Metis (1); English-speaking (all)
Marital status	Single (6): Never married (1); Married x1, widowed (1); Previously married ≥2, divorced/separated (3)
Children	Yes (3); Contact with children (1)
Education, highest attained	Less than high school (2)High school (2)Post-secondary (2): college (1); university (1)
Employment	Current (0); Past: labour (1), trades (1), service (2), business (1), sex work (1)
Sources of income	Government (6): Provincial (5), Federal (3), Private pension (1)
Average duration of six initial interviews, (range)Average duration of two follow-up interviews, (range)	77.5 (37.5–100) minutes30.5 (27–34) minutes

Canadian Observatory on Homelessness (2012) definition of “homeless”-situation of an individual or family without stable, permanent, appropriate housing, or the immediate prospect, means and ability of acquiring it. https://www.homelesshub.ca/sites/default/files/COHhomelessdefinition.pdf.

Canadian Study Healthy Aging (CSHA) Clinical Frailty Score. Rockwood K, Song X, MacKnight C, Bergman H, Hogan D, McDowell I, Mitnitski A. “A global clinical measure of fitness and frailty in elderly people”. CMAJ. 2005;173(5):489–95.

TABLE 2b

Staff demographics and interview duration

Staff Participants (N=9)
Average age (range)	51.7 (36–60) years
Gender	4 women, 5 men
Time in role at Harbour House	8.8 (3–12) years
Role description	Health Care Staff: health care aid (1) registered nurse (1) nurse practitioner (1)	Social Service Staff: frontline support worker (3) social worker (1) manager/administrator (2)
Average duration of interviews (range)	67 (43–104) minutes

It was found that HR policy fostered interactions in which 1) residents felt respected and had a sense of belonging at Harbour House; and 2) staff endeavoured to earn and maintain residents’ trust. This was facilitated by staff working closely with residents to create a care environment which helped mitigate harm from negative prior experiences, leading to improved housing stability and satisfaction (Figure 1). Data from residents and staff are presented separately below, followed by a discussion of how HR policy reduced harms related to substance use at Harbour House.

FIGURE 1

Impact of harm reduction policy in aged care—creation of a unique micro-environment at Harbour House

Residents

Residents Feeling Respected and a Sense of Belonging at Harbour House

Three residents voiced feelings of being overlooked prior to coming to Harbour House:

“We’ve worked all our damn lives and found ourselves on the street and go ‘What the hell! Is this it?’ I just want it recognized that we are a different segment of the homeless population and I think that we deserve, you know, a little bit more attention than your youngster on the street.” (female resident)

Residents contrasted their experiences at Harbour House with those in settings from which they’d been evicted or excluded in the past, often due to substance-use related behaviours.

“If you make a mistake, you don’t get kicked out. They give you a chance [at Harbour House]. [The social worker] is a very helping man. He helps you set your life up.” (female resident)

All residents voiced appreciation for the supports and freedoms at HH. For some, access to an indoor smoking area was especially important.

“We really appreciate the indoor smoke room here … We want to be able to smoke in our house … [Harbour House] sort of offers it all, I have like my privacy in my own room, access to telephone, the computer, everything.” (female resident)

Residents confirmed feeling respected, trusting it was safe to voice their ideas and express their preferences.

“We don’t see eye to eye quite a few times but at least they [staff] do respect, you know, so I would put that as probably number one priority. From time to time I’ve got myself in trouble here and … they just respect what I say.” (male resident)

Five of the resident participants described having friends in the building and appreciated having the physical space and opportunities to socialize and entertain on their own terms.

“I have several friends in this building that I see on a regular basis … we use my room because I’m set up for it. My TV is a 44-inch TV and, you know, well I just try to make it as comfortable as you can get for having a 150 square foot house.” (male resident)

Many residents identified Harbour House as home, where they felt respected and experienced a sense of belonging.

Staff

HR policy empowered staff to engage in two critical relational processes: first, working to earn residents’ trust, and then secondly, tailoring individualized care plans to maintain that trust.

Expressing Respect & Earning Trust

Staff were aware of many residents’ prior negative experiences related to homelessness, poverty, and complex health status, including substance use. They described how earning trust was key to working together and must be done with each individual resident:

“A lot of these folks with their mental health and their addiction issues have been on the fringes of society and … let down so many times along the way that you’re just another person that’s coming into their lives. I need to spend quite a bit of quality time, just even visiting with them, sitting down.” (Health care staff)

By getting to know residents’ individual needs and preferences, staff avoided making assumptions that could impair trust-building. For example, rather than assuming isolation is unhealthy and needs intervention, staff respected residents’ choices based on their knowledge of each resident.

“You have to take into account the personality of the tenant, and that some tenants become overwhelmed if they are in a group.” (Social support staff)

Staff identified strategies key to trust-building with residents, including interacting frequently with residents, and being consistent and reliable. Staff identified the importance of frequent brief encounters with residents in getting to know and trust one another, supported by physical structures such as the common dining area and centralized storage/dispensing of tobacco and alcohol.

Visiting health-care staff remarked on the value of support workers’ keen daily observations and knowledge of residents; they routinely recognized subtle but important changes in residents’ behaviours that could be early signs of physical or mental health deterioration. Continuity in the team of visiting health-care staff also facilitated trust. These health workers were very committed:

“Being there, being consistent, showing up, and honouring your word if you make a promise and you’re going to do something …” (Health care staff)

Maintaining Trust: Sharing Responsibilities & Individualizing Supports

Staff committed to individualizing care plans. This included voluntary managed alcohol and tobacco supports for those with severe dependence for whom abstinence was not possible. For some residents, this involved centralized substance storage and timed daily dispensing, facilitating frequent face-to-face interactions, and opportunities for rapport development. For others, only periodic supports were provided. Supports were based on each resident’s needs and preferences, regularly negotiated with the social support staff:

“For some we dispense it [alcohol] from the main office. It’s a different level of involvement for, you know, you work with each individual tenant with what they need at that time … some just go for long periods of time without drinking. They just binge drink … They want us just to go and remove the alcohol if they have it.” (Social support staff)

Social support staff indicated approximately half of those on managed alcohol significantly reduced their use of their own volition over time. Many eventually self-manage, and staff commented on how this contributes to enhanced wellness as evidenced by their hygiene and social interactions with others.

While a few staff noted that ‘others’ (past providers or other current providers) were reluctant to administer alcohol to people with an existing substance use disorder as a potential additional harm, no participants owned a personal reluctance. This concern was one of several external (meso- and macro-level) constraints on staff’s trust-building/maintaining efforts. These included funding at the level of a shelter rather than supportive housing model, having to rely on for-profit vendors for frontline services who do not subscribe to HR practices, and being over-subscribed due to being a single site offering HR for older adults in a large urban centre. Though infrequent, such inconsistencies between staff and additional external constraints on trust-building highlighted the need for concentrated and sustained efforts to build and maintain trust.

Residents & Staff Working Together—Trust Defusing Tensions

There was tension, at times arising from staff’s efforts to balance individuals’ preferences with the supports needed to maintain housing or personal safety. One 78-year-old participant had moved to Harbour House after being evicted at age 70 from her seniors’ apartment for behaviours related to severe alcohol dependence. At Harbour House, despite receiving 3 oz of vodka every 4 hours under her managed alcohol agreement, she periodically sought additional alcohol. This was a common experience amongst managed alcohol program (MAP) participants and contributed to physical harm, including repeated falls and hospitalization. Rather than relying solely on this resident to change her behaviour, staff focused on making the environment safer by increasing monitoring and removing excess supply. Although frustrated by staff’s intrusion, she trusted the staff and recognized she needed their support.

“A while ago whenever I went out, when I would come back in, they used to search my handbag for alcohol and that really bugged me but they don’t do it anymore … I honestly think I might be tempted to overdo it, you know [without the staff’s help].” (female resident)

DISCUSSION

Few studies have considered the experiences of older people with homeless histories and their care providers in housing grounded in harm reduction (HR). The concept of risk environments helped reveal how HR policy at Harbour House helped staff to work closely with residents, providing structurally competent care( which offset harms related not only to substance misuse but also to prior experiences of neglect and rejection, and associated distrust of care providers. Trusting relationships were evident amongst staff and residents, which helped to foster stabilization and a sense of home, and were the product of the policy, as well as the physical and social environments at Harbour House. Similar benefits have been seen in studies of younger people (average age 42) in smaller scale HR housing (15 beds) with mandated managed alcohol supports; participants reported feeling empowered to make positive change, including making friends and reconnecting with family.( Increased quality of life, sustained housing stability, and cost-savings were also evidenced in highly resourced Australian models of HR housing for older people living with alcohol-related brain injury.(

The approach by staff and the suite of services offered at Harbour House helped reduce previously identified barriers to care and the stigmatization that older people experiencing homelessness face: long waits to be seen, minimum required age (65 years), high costs of home care and transportation, and fears of being treated poorly by physicians.( In the face of lost trust and sense of control while homeless, the strong need for trusting relationships and self-determination is well-documented, including in a survey by McDonald et al.( of 257 older adults (average age 57) in Toronto and Calgary.( Staff recognized residents’ need for trust and autonomy, accepting the reasons for ‘risk behaviours’, such as self-isolation and continued substance use, as legitimate and potentially ameliorated through the environment at Harbour House. In a Washington state study of HR housing, participants (average age 48.3) reported heavy alcohol use to self-medicate psychiatric symptoms and to feel a sense of belonging.( In this “project-based” single-site setting with multiple independent units, Collins et al.( found that frequent contact with staff allowed “micro-interventions”—check-ins with residents regarding their physical and social health/wellness. Similar frequent and helpful interactions between staff and residents were supported by the physical design and HR program structures at Harbour House. Physical and social proximity allowed staff to develop deep understanding of residents’ preferences and needs, both related and unrelated to substance use. This understanding inspired staff to modify the environment (e.g., remove excess alcohol, express regard for residents) and collaboratively reduce harm related to risk behaviours, rather than rely solely on individual residents to change their behaviour.( Residents reported feeling respected, and accepted unrequested (and sometimes frustrating) supports because of established trust and relationships with staff. Similar routine interactions in a Norwegian residential drug rehabilitation facility fostered younger residents’ sense of self-worth and confidence in staff’s intentions.(

Harbour House staff framed risk in terms of potential threat to residents’ physical health and housing stability, though scholars have noted possibilities for additional risks to staff when limiting alcohol to people with alcohol use disorder.( Recent study of homeless sector workers revealed rates of post-traumatic stress disorder (PTSD) symptoms significantly greater than rates seen in frontline staff providing emergency services.( Escalating violence towards shelter staff and workplace safety concerns are driving workers in Calgary, Alberta to unionize, and is an area warranting further study.(

Integration of HR into housing for people experiencing homelessness is not new. For more than 20 years, “Housing First”, a rights-based approach which is not contingent on housing readiness or compliance with treatment, has dominated North American homeless policy and shown considerable benefits for younger adults.( However, prevailing Housing First approaches may require adaptations to yield fruitful environments for HR among homeless older adults. For example, Housing First favours independent units, “scattered” amongst market-housing, relying heavily on off-site supports, or periodically visiting supports that are time-limited and emphasize a return to independence.( Similar to abstinence-only policy in aged care, Canadian homeless policy largely overlooks the needs of older people.( With significant physical, cognitive, and/or mental health challenges, the suitability of scattered site Housing First for older people has been questioned.( A spectrum of models in North America are needed to meet the diverse needs of the growing older homeless population, and the congregate model of supportive housing with HR at Harbour House could help to address significant gaps in the current aged care landscape.(

Limitations

The study was conducted in a supportive housing facility in Western Canada, as such the findings and implications may not be transferable to other contexts (e.g., larger supportive housing complexes, in different countries). However, the findings’ alignment with broader theory in harm reduction principles (choice, dignity, autonomy)( and ‘Rhodes’ risk environment framework’( indicate potential for transferability to similar contexts.( The study was limited by the small sample, and resident participants were predominantly of European/settler ancestry. Further study is needed of demographic subgroups based on ethnicity and gender. There was also the possibility that some participants’ reports were influenced by staff’s ongoing employment and residents’ prior relationship with a principal investigator. Further study is needed to explore staff’s perceptions of risk, ideally through longitudinal mixed-method approaches to encourage a sense of safety to disclose. Several measures were taken to mitigate potential negative impact on participants’ reports related to prior relationship with a principal investigator, including giving participants the choice of their interviewer; it could be argued that the investigator’s ongoing work as a clinician at Harbour House may foster greater safety and trust, while also providing opportunities for participants to hold her more accountable for the findings.

CONCLUSION

There are people for whom complete abstinence is an appropriate approach to the delivery of long-term housing options, and others for whom harm reduction is best. Health policy needs both, and needs to be provided with evidence for identifying which individuals would benefit from one or the other. Our study shows the value of integrating HR policy in housing-based care of older people with health and social complexity related to homelessness. Building trusting relationships and managing competing priorities (and risks) are cornerstones of geriatric care. Both are facilitated by adoption of harm reduction policy, enacted at the level of the physical and social care environment which, if more widely adopted, could improve the care and dignity of this small, but growing, older community.