Key challenges in providing services to people who use drugs: The perspectives of people working in emergency departments and shelters in Atlantic Canada

Abstract

Aims: Many people who use drugs (PWUD) have multiple health and social needs, and research suggests that this population is increasingly accessing emergency departments (EDs) and shelters for health care and housing. This qualitative study explored the practices of those working in EDs and shelters when providing services to PWUD, with a particular focus on key challenges in service provision. Methods: EDs and shelters were conceptualized as ‘micro environments’ with various components (i.e. social, physical and resource). One-on-one interviews were conducted with 57 individuals working in EDs and shelters in Atlantic Canada. Findings: The social, physical and resource environments within some EDs and shelters are key forces in shaping the challenges facing those providing services. For example, the social environments within these settings are focused on acute health care in the case of EDs, and housing in the case of shelters. These mandates do not encompass the complex needs of many PWUD. Resource issues within the wider community (e.g. limited drug treatment spaces) further contribute to the challenges. Conclusions: Structural issues, internal and external to EDs and shelters need to be addressed to reduce the challenges facing many who work in these settings when providing services to PWUD.

Background

In Canada, many people who use drugs (PWUD) (individuals who use illicit drugs and/or licit drugs other than medically intended) have been found to have multiple physical and/or mental health concerns (Fischer et al., 2005), and rates of premature morbidity are higher among PWUD than non-PWUD populations of the same age (Popova, Rehm, & Fischer, 2006). The health of PWUD is ‘inextricably bound to their social environment,’ and is a ‘product of both drug-use behaviours and social determinants’ (Galea & Vlahov, 2002, p. S136). Indeed, experiencing poverty, homelessness, incarceration or other forms of social exclusion have been found to be powerful determinants of drug users’ health and death rates, and also influence risky drug use behaviours (e.g. sharing drug using paraphernalia) (Galea & Vlahov, 2002; Najman, Toloo, & Williams, 2008).

Evidence suggests that social factors also influence the types of services PWUD access, and the frequency with which they access them. Studies, including Canadian studies, have found, for example, that PWUD seeking health care are more likely to do so via emergency departments (EDs) rather than primary care services (Hansagi, Engdahl, & Romelsjö, 2012; Kerr et al., 2005; Palepu et al., 2001), with the unstably housed making relatively frequent visits to EDs (Fairbairn et al., 2011; Haber, Demirkol, Lange, & Murnion, 2009; Palepu et al., 1999, 2001). These findings have been interpreted to mean that the most socially marginalized PWUD, such as those who are homeless, are not only more likely to suffer poorer health, but that they are also less able to establish on-going relationships with health care providers and/or address health concerns before they become acute (Palepu et al., 1999).

The social marginalization of PWUD is taking place in a context of systematic cut-backs to both health and welfare spending. Since the 1980s, cutbacks have been occurring within Canada under the auspices of deficit reduction and neo-liberal policy reforms (Bezanson, 2006; Katz, 2003; Kitchen & Curry-Stevens, 2006; Lightman, Mitchell, & Wilson, 2008; Lightman & Riches, 2000; Rice & Prince, 2000). Various ‘cradle to grave’ benefits and services (Katz, 2003, p. 29)—including family allowances, old age security and unemployment insurance—have been ‘claw[ed] back’ by the Canadian government (Katz, 2003, p. 28). During this period of time, changes in the health care system have also led to increasing wait times for health care across the country (Armstrong & Armstrong, 2010; Walberg & Björnberg, 2008). In addition, numerous essential social services, such as housing, have to a great extent been downloaded to non-profit, community-based organizations which are now serving an increasing number of homeless or under-housed people through the shelter system (Evans, 2011; Kosny & Eakin, 2008; Muckle & Turnbull, 2006).

It is within this context of the erosion of the social safety net in Canada (Rice & Prince, 2000), and increasing numbers of PWUD accessing EDs and shelters that we conducted our study. A key goal of our research was to understand some of the current practices within EDs and shelters in terms of the provision of care to PWUD. We were specifically interested in understanding how the ED and shelter environments shape the practices of people who work in these settings, including any challenges they face in providing services to PWUD. There is a fairly substantial body of literature examining the challenges PWUD face when accessing services (e.g. stigmatization), especially health care services (Neale & Kennedy, 2002; Parker, Jackson, Dykeman, Gahagan, & Karabanow, 2012; Spence et al., 2008). However, less is known about the challenges faced by those who work in EDs and shelters when providing services to this population. These challenges are the focus of this paper.

We recognize that EDs and shelters offer two very different types of services, with EDs focused on providing emergency health care, and shelters centred on housing. Moreover, it is acknowledged that shelters tend to be organized in a more informal manner than EDs (Karabanow, 2002). Nevertheless, for this study, we were interested in exploring whether or not there might be some similar challenges in providing services to PWUD across these two types of settings. Understanding potentially similar challenges was of interest because it points to the possibility of inter-sectoral collaboration as a means of addressing the challenges. In addition, it is clear that at risk marginalized populations often use both systems in their day-to-day survival.

We conceptualized EDs and shelters as ‘micro environments’ following the perspective outlined by Tim Rhodes (2002, 2009; Rhodes & Simic, 2005). By focusing on these micro environments, we shift how we think about issues ‘from constituting individuals alone’ to ‘socio-political situations, and structures in which individuals find themselves’ (Rhodes, 2002, p. 88). This focus allows for an exploration of how the practices of those working in EDs and shelters are shaped by the context within which they work. Moreover, it draws attention to the structural changes that are needed to improve the services and supports for PWUD—thus moving beyond individual-level change. In this paper, we specifically examine the challenges to providing services to PWUD in terms of three key components of the ED and shelter environments: the social, physical and resource components. An examination of the policy component (as per the Rhodes’ framework) is beyond the scope of this paper.

Methods

Prior to data collection, ethics approval was obtained from the relevant university and hospital institutional ethics review boards in the four Atlantic Canadian provinces where the study took place: Newfoundland and Labrador, Nova Scotia, New Brunswick and Prince Edward Island. One-on-one, semi-structured, qualitative interviews were conducted with 57 individuals, 43 of whom work in EDs and 14 in shelters. The sample was composed of 16 men and 41 women, and all individuals interviewed provided voluntary, informed consent prior to the interview. Fifty-two interviews were conducted face-to-face and five on the telephone. In order to try and capture a diversity of experiences, interviews were conducted in EDs and shelters in both urban centres as well as outside of urban centres in rural or semi-rural communities. Interviews began in January 2011 and were completed in May 2012.

The total population of the four Atlantic Canadian provinces is approximately 2.3 million (Statistics Canada, 2013). Based on a 2011 Health Canada survey (2012), lifetime use of any five drugs (cocaine/crack, speed, ecstasy, hallucinogens and heroin) is as follows: 10.0% for Newfoundland & Labrador; 10.1% for Prince Edward Island; 11.9% for Nova Scotia; and 10.4% for New Brunswick (Health Canada, 2012). Lifetime use for cocaine/crack alone is as follows: 4.4% for Newfoundland and Labrador; 3.8% for Prince Edward Island; 4.4% for Nova Scotia; and 3.7% for New Brunswick (Health Canada, 2012).

Recruitment and interviewing

Individuals were eligible to participate in this study if they were a frontline staff, manager, or administrator in an ED or shelter in Atlantic Canada, and could speak to practices and policies related to providing services to PWUD. Individuals were recruited through notices and posters distributed within EDs and shelters that agreed to participate in the study. The research co-ordinator conducted all of the interviews. Interviews were audiotaped with the permission of participants, otherwise notes were taken and subsequently typed (n = 11). All participants were offered a small honorarium to thank them for their participation.

Audiotaped interviews were transcribed by a professional transcriptionist with names or other personally identifying information removed. Transcribed interviews were entered into a qualitative software program (Atlas.ti) to assist with data management and analysis. Participants wishing to review their transcript were sent the transcribed interview.

Interview guide and data analysis

Semi-structured interview guides were utilized to collect the data. The guides were developed by the research team composed of researchers (with years of experience in the areas of addictions, harm reduction and housing), as well as physicians and workers in community agencies. The interview guides sought to understand what typically happens when PWUD go to the ED or shelter, why certain practices might be followed (e.g. is there a formal or informal policy that informs the practice?), as well as any challenges in providing care or services to PWUD, and any existing or suggested practices or policies for addressing challenges.

The first few interviews in both EDs and shelters were read and re-read by the research co-ordinator and the principal investigator to determine key concepts within and across the interviews (e.g. resource constraints and safety). Evolving concepts were discussed with other team members until there was agreement, and the transcripts were coded based on the coding structure. All data were read and re-read multiple times, and relationships between themes and sub-themes noted and discussed with members of the research team. Following the process outlined by Corbin and Strauss (2008), the evolving patterns were thought and re-thought through the constant comparison of concepts and ideas, and a reading and re-reading of the data. This process continued until there was a full integration of all concepts.

Limitations

A key limitation of this study is that the data are based on a convenience sample of people working in EDs and shelters across Atlantic Canada. As such, the study results cannot be interpreted to represent all the challenges, issues, or perspectives of people working in EDs and shelters within the region. Rather, the results provide an understanding of some key issues based on the perspectives of people working in a select number of EDs and shelters. Moreover, our study was focused on shelters for adults (including shelters that house women and their children), and therefore, our data do not include challenges that might specifically face individuals working in shelters servicing youth.

It should also be noted that we interviewed more people working in EDs than in shelters. A greater number of people were available to be interviewed within EDs than in shelters as many of the shelters are staffed by only a few individuals, and there are relatively few shelters outside of major metropolitan centres. Additionally, we did not ask participants about specific drugs they believed PWUD were taking. Although the PWUD population might include those who only use cannabis, it is likely that when speaking about this population, participants were referring mainly to individuals who have significant substance use issues, and require particular attention.

Results

Our findings reveal several key challenges to providing services to PWUD within the ED and shelter environments. However, it is important to note, that not all of the challenges apply to each ED or shelter because of the diversity across EDs and shelters in terms of their physical size, location (e.g. rural or urban), specific workplace policies, etc. In particular, shelters vary in terms of their policies about who can stay. In some shelters, PWUD can stay as long as they are not actively using drugs, and in other shelters the policy about who can stay centres around safety issues (e.g. is it safe for the person to be at the shelter?).

In keeping with our conceptual framework the findings are organized in terms of three key components of ED and shelter environments: (1) the social environment or social context within which staff and PWUD interact; (2) the physical space; and (3) the resource environment. These three components are clearly interconnected as the social environment can, at times, be influenced by the physical space (e.g. physical spaces can affect social interactions). Likewise, the resources available to EDs and shelters can affect the social environment, including how quickly patients can be seen by specialists within an ED, and the number of clients who can be housed at a shelter.

The social environment

Service provision in a context of PWUDs’ multiple and complex needs

EDs and shelters both have specific mandates. EDs have a mandate to provide emergency care, and shelters have a mandate to provide emergency and/or temporary housing. However, individuals going to EDs and shelters, including PWUD, often have multiple needs beyond the mandates of these two organizations or agencies. For example, individuals sometimes go to the ED not only for an acute health issue, but a variety of chronic health problems or ‘non-emergency’ complaints and concerns. A number of ED staff who participated in this study indicated that PWUD may access an ED because they have an acute addiction-related health problem (e.g. overdose), and/or because they have another non-acute health concern which might include a mental health issue. Likewise, a number of shelter staff indicated that sometimes PWUD not only need housing, but legal assistance and/or support and services for significant addictions and/or mental health problems. For some individuals working in both EDs and shelters, providing services for PWUD with these multiple needs is challenging.

Within the ED environment, challenges related to non-acute health issues often centre on the significant time required to address these issues. As noted by a few participants, the environment within an ED is focused on providing a medical assessment and service relatively quickly, and this stands in contrast to what is often required in addressing a chronic health issue such as a mental health concern, or an addiction issue. One participant working in an ED noted that if someone has a mental health concern, and the mental health nurse working in their ED is available, the person would be called. Sometimes, however, the mental health nurse is not available and in such instances it can be difficult to find the time for the patient.

… I find that we don’t have time for mental health patients or someone seeking addiction services. If they come in and they’re under the influence [of drugs], we have to deal with the [urgent] medical problems of the influence first. [A17ED]

Some shelter staff also noted that the multiple and complex needs of PWUD, beyond housing, can present a real challenge. One shelter staff participant pointed out that staff sometimes feel ‘ill-equipped’ to deal effectively with these complex issues, and commented that the number of clients presenting with addictions and/or mental health concerns has increased in the recent past.

Drugs, alcohol, mental health are a big part of what we deal with now [more] than 30 years ago … from a staff perspective, our front-line staff are extremely challenged. They often feel out of their element … So staff often feel ill-equipped to deal with a lot of the issues that they’re being faced with. [A18S]

This participant also described how the time required to help PWUD, and refer them to a social worker or mental health counsellor, is not always available as it ‘depends on what's happening at the time [in the shelter]’ including the number of staff on duty [A18S].

Some participants in both EDs and shelters also commented on the challenges associated with PWUD who may be frustrated or angry, and whose behaviour or language may be inappropriate. A few ED participants indicated that often PWUD come to the ED, and are cared for without any problems. However, at times interactions between PWUD and staff are problematic because of tensions and frustrations. One area of tension is the disconnect between the expectations of PWUD, and what some ED staff feel they can provide, or believe is appropriate, in the ED context. For example, when PWUD go to the ED to refill a narcotics prescription, some health care providers are reluctant to refill the prescription because they believe it is the role of the family physician and not the ED physician to refill prescriptions, or they may question whether or not the prescription will be used for what the health care provider views as a legitimate medical problem [A44ED]. Participants working in EDs also indicated that sometimes there are frustrations when PWUD return on a regular basis to the ED for non-emergency issues, and that they may be returning because they do not have a family physician from whom they can obtain primary health care [A34ED]. In some instances, PWUD experience long wait times because patients with acute health problems have to be seen first, and PWUD may become frustrated and engage in ‘venomous’ talk [A1ED]. One ED participant argued that although one needs to treat all patients the same, including PWUD, and staff try to do this, it can be difficult in the case of ‘frequent fliers’ or those who return frequently to the ED. Yet another ED participant expressed frustration over the fact that some PWUD are looking for health care in an ED yet their basic needs have not been met. For this individual what some PWUD need is ‘a place to live and food, cleanliness’ [A34].

Challenges related to the behaviour of some shelter residents were also commented on by some shelter participants. One shelter staff participant, for example, noted that shelter work can be especially challenging when single-staffed (often during the overnight shift) and there is ‘escalating behaviour’ on the part of someone in the shelter. Although behavioural problems are not exclusive to PWUD, there are clearly challenges in responding to some behaviours. ‘You’re single staffed so you’re trying to control everybody else in the shelter and now escalating behaviour. So that sometimes can be a challenge’ [A14S].

A couple of shelter participants also talked about issues associated with managing relationships or interactions between shelter residents, and specifically behaviours on the part of PWUD that might be viewed negatively by other residents [A41S] or act as a ‘trigger’ for someone's addiction. One person commented, ‘… like [if] someone [is] passing out in their dinner or something… other residents… can sometimes feel a little distressed by that or a bit judgmental’ [A2S]. Another participant noted that, ‘and some things, especially if it's the smell of alcohol or if it's something else, that it may be a trigger smell for somebody else that's sharing the room’ [A19S].

Ensuring safety

The safety of everyone in the ED and shelter environments was a common theme we heard from a number of participants. In both settings, participants spoke about various safety policies and practices that exist to ensure everyone's safety. The practices and policies centre on physical safety issues, and within the ED setting include the proper disposal of needles, designated spaces for patients who are physically aggressive, or calling security or the police if one is concerned about their safety or the safety of others. One participant argued that in some cases, speaking calmly to patients is a practice used in response to individuals who are agitated to help ‘de-escalate them’ [A37ED]. However, another participant commented that the potential for violence and feeling unsafe is a challenge in providing care for PWUD because their behaviour can be unpredictable due to their drug use and/or mental health issues.

Because there are some situations sometimes that you feel unsafe in dealing with some of these people because you’re not really sure of their behaviour and how it's going to be … sometimes we have people who try to … they’re kicking and punching and biting and spitting …. [A15ED]

A number of shelter staff also discussed the importance of ensuring the physical safety and security of individuals within the shelter – all residents and staff [A9S; A18S]. The specific safety policies and practices varied across shelters but included locking doors at night so no one can enter after a certain hour (unless under certain circumstances), and adding additional staff if there are particular safety concerns. The importance of everyone's safety was clearly articulated by one shelter staff member who commented that, ‘I have my own self that I need to protect, and I have all these other people that I need to protect as well’ [A18S]. When speaking about the safety of PWUD, another shelter worker argued that it was concerning when residents were not forthcoming about their drug use because it was unclear what to tell health professionals if there was an emergency, and paramedics had to be called.

Well, a lot of times some of the challenges are, are they being honest with what they’re taking? … And for us, it's not a point of judging you for what you’re taking, it's we need to know in case something happens and the paramedics come and we can tell them. This information can save somebody's life. [A14S]

The physical space

The limits of the physical space

For some individuals working in both EDs and shelters, challenges related to the physical space within their setting were also noted. Although some participants pointed out that within the ED where they worked there were enough spaces for private, confidential discussions with patients, one participant indicated that in their ED this was not the case if the person was thought to be a safety risk. In such instances, any confidential discussions took place at the person's bedside, with the curtain drawn and speaking as quietly as possible [A23ED]. Another participant commented that having private, confidential conversations with PWUD was definitely a challenge in the ED where they work. According to this person, ‘But actually inside the [emergency] department, there's nowhere private… there's not usually anywhere. A curtain, that's about it’ [A16ED].

A couple of ED participants also indicated that PWUD sometimes wait in the ED space for long periods of time thus ‘taking up’ limited space. In some instances, they are waiting to be ‘medically cleared’ or for a referral to another service. As one participant commented with respect to PWUD with mental health concerns, there is sometimes a long wait in the ED before they can be seen by psychiatry, particularly during the overnight shift, which places a ‘burden’ on the ED. As this participant noted, ‘… we [sometimes] hold them [PWUD] overnight to be seen by psychiatry in the morning. Which has actually placed a bit of burden on the emergency room’ [A27ED].

Although individuals in shelters did not speak about issues related to having confidential conversations or long wait periods for individuals to be admitted to the shelter, the limits of the physical space in some shelters was noted. One shelter participant spoke, for example, of having to turn people away at times when the shelter had insufficient space [A25S]. Another shelter participant indicated that the physical configuration of the space can represent a challenge in terms of housing PWUD. For example, stairs can represent a safety hazard for individuals under the influence of drugs and/or alcohol.

Few housing spaces for PWUD

Individuals working in both EDs and shelters commented on the lack of housing options for some PWUD, and the fact that EDs and police stations are sometimes the only place for PWUD to go to be safe and warm overnight. One ED participant commented that, ‘… Most times we will house them here for the night [if no place for them to go]. If they’re coming in through the evening or overnight, we’ll house them here’ [A26ED].

Likewise, a few individuals working in shelters spoke of the limited housing options for this population particularly if they were actively using drugs, and could not stay at a shelter for safety reasons. As one participant mentioned, some family members of PWUD will not provide them with housing because of their drug use [A19S], so PWUD have few options if they cannot stay at a shelter. At times, PWUD might go to a hotel, or a local ED, or to a local police station. ‘The local police are very good… They, like [us], don’t want to see no one sleep outdoors… and they will just take them [PWUD] and put them in, make sure they have a bed, if need be’ [A42S].

Another shelter worker commented that because of the few housing options for this population, ‘… you’re forcing people onto the streets, into places like the emergency rooms and the drunk tanks [police station] and so on because there is nowhere else for them to go. And it's a real challenge here’ [A18]. When speaking about ‘drunk tanks’ this person was referring to police detaining someone who is highly intoxicated by placing them in a jail cell (or separate facility at the police station) in order to ensure their safety, and/or the safety of the public. Commenting on the difficulties for this population in finding housing, another participant suggested that, ‘Well a huge need is supportive housing’ [A9S].

Limited resources

In-house and in the community

Some ED staff reportedly have limited addictions and/or mental health training (as their training is in acute care), although some have, over the years, obtained extensive experience in working with PWUD. In some EDs, specialized staff (e.g. social workers and psychiatric nurses) are available during weekday shifts, but it can be difficult to access these specialists during overnight shifts and/or on the weekends.

So the time when we need the [crisis team] is usually at 3:00 in the morning on Sunday or Saturday night, and they don’t come to work until Monday. So I think it's about having the right resources at the right time. [A5ED]

A few participants working in shelters echoed some of these same challenges indicating that they too have limited staff with specialized training to work with people with addictions and/or mental health issues. One person also mentioned the limited budget available to provide needed training to shelter staff on how to work with people with addictions and/or mental health issues [A18S].

In addition to the limited availability of mental health and/or addiction specialists in some EDs and shelters, a few participants also spoke about limited drug treatment spaces within their community [A24S]. One ED participant specifically mentioned challenges related to referrals for methadone maintenance treatment (MMT) because of long waitlists. ‘Now that's [methadone maintenance treatment] something that we have rather limited… we have limited access to, if at all. There's only a very few methadone prescribers [in community]’ [A27ED].

Another ED participant commented on the great frustration that they felt when a patient was ready for detoxification, and yet there were no spaces in the program for the patient [A29ED]. Cumbersome admission criteria for such programs in their community were also noted as patients have to call each morning to inquire about openings, ‘and if they don’t call every morning they fall off the list’ [A29ED].

Conclusions and discussion

The macro environment within which service providers in Atlantic Canada work is one where there have been reductions in health and social spending over the past 20–30 years, and where there has been an erosion of the public safety net for many living in poor socio-economic conditions (Muckle & Turnbull, 2006; Raphael, 2011). Informal, community-based services, such as shelters, have developed to provide for those in need, and although these services were originally established as temporary services they have become ‘permanent features of the social service landscape’ (Kosny & Eakin, 2008, p. 149). In addition, many PWUD access shelters because they cannot afford stable housing or because they are ‘unwelcomed’ at the homes of family (Glasser & Zywiak, 2003). Some PWUD not only ‘end up’ at shelters but also EDs, sometimes on a frequent basis, to obtain care for chronic non-emergency health concerns including mental health and addictions issues. The reasons why they utilize EDs for non-emergency issues vary but include a lack of ready access to primary health care (Fairbairn et al., 2011; Kerr et al., 2005; Morris & Gordon, 2006; Schwarz, Zelenev, Bruce, & Altice, 2012).

Several studies of PWUD's experiences accessing health care highlight negative practices on the part of some health care professionals (Beyon, Roe, Duffy, & Pickering, 2009; Butters & Erickson, 2003; Neale, Tompkins, & Sheard, 2008). However, with a few notable exceptions (e.g. Pauly, 2008; Paterson, Hirsch, & Andres, 2013), there is relatively little research exploring the challenges for health care providers particularly in terms of how the ED environment influences their practices. This type of research is important because, as Paterson et al. note, one cannot simply ‘blame practitioners for bad behaviour’ as this ‘negates the powerful influence of the structures both within and external to the ED’ that play a role in influencing such behaviours (2013, p. 477).

The research by Henderson, Stacey, and Dohan (2008) clearly demonstrates the importance of taking heed of structural forces as these researchers found that even when health care providers working in EDs held positive attitudes towards PWUD, stigma, at times, prevailed. According to these researchers, such issues as resource constraints within the ED can contribute to stigmatizing attitudes and practices when, for example, PWUD are perceived as wasting resources (e.g. returning to the ED too frequently) (Henderson, Stacey, & Dohan, 2008).

As our research found, and as others have reported, the focus of EDs is on ‘fixing’ acute physical health problems quickly (Marynowski-Traczyk & Broadbent, 2011; Pauly, 2008). As a result, within this social environment there is relatively little time to provide the often extensive support and care that is needed to address chronic addictions and mental health issues. In addition, there are sometimes resource constraints as health care professionals with training and expertise in addictions and mental health are not always available, especially at night. This means that PWUD sometimes have to wait long periods of time in the ED for referrals to specialized care, or because their concerns are non-acute, and acute care takes priority. The challenges of wait times and a lack of available expertise have also been reported by Paterson, Hirsch, and Andres (2013) in their research involving PWUD who are Hepatitis C positive.

Our research further found that how the physical space is configured in an ED can create challenges to having a confidential conversation with patients, including PWUD, and this physical design problem was also described by Paterson et al. (2013). A lack of space for confidential conversations can affect what information PWUD might provide given the stigmatizing nature of drug use.

Not only do social, resource and physical structural issues create challenges for staff working in EDs, but also for those working in shelters. Many shelters provide such services as counselling and meals, but their principle service is housing, and the social context is one that is focused on providing shelter for those in need. Within some shelters, staff have the training and/or experience to help PWUD with a number of their concerns. However, in other shelters these resources do not exist or are limited as the staff expertise and time to respond to PWUD's multiple needs is constrained, especially during overnight shifts when the needs of others have to be managed. In addition, the physical space available within a shelter may mean that sometimes potential clients, including PWUD, have to be turned away because the shelter is full.

One strategy for responding to some of the challenges identified within the social, physical, and resource environments of both EDs and shelters is to increase the number of available staff, as well as to enhance expertise/training in addictions and mental health, in places where this is a concern. This would allow the needed time for PWUD, and help to ensure they receive the supports and services that they require in a timely manner. Another strategy would be to alter the space configuration to ensure confidential discussions in EDs (where this does not currently exist) or ensure shelters have ample physical space to house PWUD who require shelter.

Although it would be of value to address components of the social, physical, and resource environments within EDs and shelters that are shaping key challenges, our research, as well as that of others (Paterson et al., 2013), indicates that external issues also need to be addressed. Specifically, our research found three key external challenges: PWUD's lack of access to primary care; the limited number of drug treatment spaces or, in some places, difficulties obtaining access even when available; and, the relatively few housing options, particularly when PWUD are unable to stay at shelters due to safety concerns and/or because of their drug use (in cases where shelter policies prohibit those under the influence from entering).

The particular strategy or strategies for responding to PWUD's lack of access to primary care will likely vary from community to community depending on the size of the population of PWUD. However, Islam, Day, and Conigrave (2010) and Islam, Topp, Day, Dawson, and Conigrave (2012) describe primary health care centres that are specifically designed for the PWUD population, and these types of centres are one possible response because they recognize the unique and multiple health and social needs of many PWUD. According to Islam et al. (2012), the services that are available in these centres vary from place to place with some having needle-exchange programs, wound/vein care, physician/nurse consultations, infectious disease testing, vaccinations, and counselling. The majority of the sites also provide mental health services, and some provide various other social services: ‘including meals, telephone and sometimes internet facilities, rest-rooms, coffee and snacks, legal services; haircuts, and/or showers and washing facilities’ (Islam, Topp, Day, Dawson, & Conigrave, 2012, p. 96). In addition to providing primary health care, these centres could also offer drug treatment programs, and thus provide improved access to treatment for ‘hard to reach’ PWUD or those who are reluctant to access treatment centres because of the stigma associated with such centres (Radcliffe & Stevens, 2008; Stevens, Radcliffe, Sanders, & Hunt, 2008).

Our research indicates that changes are also needed in order for PWUD to have timely, and easier, access to drug treatment programs. More spaces or openings within programs where waitlists are an issue would be of value, and it would also help to re-think the admission criteria in instances where this is a deterrent to access. In some places, the integration of treatment programs into a shelter might be considered. In Dublin, Ireland, some shelters/hostels which provide emergency accommodation for homeless people have established a methadone program within the facility and methadone is delivered by general practitioners and nurses (O’Reilly & Murphy, 2011). According to an evaluation of one such shelter, a key strength is that, ‘Residents can see a health professional almost immediately on accessing the Shelter and be assessed for treatment’ (Geraghty, Harkin, & O’Reilly, 2008, p. 11). It might also be helpful in some communities to have advocates for PWUD in EDs to help navigate access to treatment and save ED staff time that could be used for direct care. In New Haven, USA, for example, ‘Health Promotion Advocates’ who work in an ED specifically assist with drug treatment referrals through a number of initiatives including the development of strong linkages with community treatment programs (D’Onofrio & Degutis, 2010).

The limited housing options for PWUD suggest that there is also a need to re-think housing for PWUD (Stax, 2003). Currently, within Canada, a ‘Housing First’ program has been adopted in a number of cities including one city in Atlantic Canada (Moncton, NB) (Pauly, Reist, Belle-Isle, & Schactman, 2013). This is a model that can potentially help to address the need for long-term, supportive, stable housing for PWUD in other locations. This Housing First program focuses on housing people regardless of current patterns of substance use, and a key principle of the program is harm reduction as ‘individuals are not required or expected to undergo treatment for substance use or to abstain in order to access and keep permanent housing’ (Pauly et al., 2013, p. 284).

The different types of programs or policies that might be implemented within any particular community in Atlantic Canada to address the varied challenges raised by our research will depend, of course, on a number of factors including economic resources. However, research indicates that people who are injection drug users often frequent EDs for non-acute health concerns, and this is costly (Sweeney, Conroy, Dwyer, & Aitken, 2009). The injection drug using population is only a proportion of the population that uses drugs, and these costs are particularly linked to the injection drug using population. However, some costs for the treatment of non-acute health concerns may also be associated with the population of drug users who are not injecting drugs, including those with chronic mental health issues.

Any changes that provide better access to primary care and thus help to prevent secondary health problems will also likely reduce costs. Stable housing should further help to save costs in the long term because PWUD will be able to reside in an environment that provides continuous housing conducive to improved health and less use of expensive public services including emergency room visits (Cities and Environment Unit, 2006; Gulcur, Stehancic, Shinn, Tsembris, & Fischer, 2003; Stax, 2003).

Regardless of the types of policies and programs that might be put into place the re-thinking and re-structuring of health and housing services for PWUD will require recognition of the diversity of the PWUD population. This population is not homogeneous but diverse in terms of gender, ethno-racial background, sexual identity, etc. This means that any intervention that fits the needs of one sector of the population may not be appropriate for another. PWUD who have children, for example, may require different types of housing accommodations than those without. It is imperative, therefore, that individuals from all sectors of the population are part of any initiative for change, and that their needs are recognized and understood in order to ensure that changes meet the needs of all within the population and not just a select few.

A word about the conceptual framework

A key objective of our study was to understand some of the challenges facing individuals who work in EDs and shelters in terms of the provision of services to PWUD. We utilized a modified version of the framework outlined by Tim Rhodes (2002, 2009; Rhodes & Simic, 2005) which was developed to explain how drug harms are shaped by ‘risk environments’. This risk environment framework was a useful heuristic device that helped us understand how service providers’ practices, including the challenges they face with providing services to PWUD, are shaped by the environment. We conceptualized EDs and shelters as ‘micro’ environments within a larger macro environment, and focused on three components of these environments (i.e., social, physical, and resource). Centring on three components allowed us to disentangle some of the complexity of the micro environments of EDs and shelters while demonstrating ‘the dynamism of reciprocal relations between environments and individuals’ (Rhodes, 2009, p. 193). Rhodes and others have highlighted the ‘“nonlinearity” of interactions… inseparability of “levels”… and daunting array of “factors”’ involved in the relationship between drug-related harm, individuals and environments (Rhodes, 2009, p. 193). Our data also highlight these issues with respect to service providers and their environments.

A key strength of Rhodes’ framework is that it helps to reveal the structural factors that shape the contexts in which individuals, like service providers, find themselves. This in turn points to organizational changes necessary to meet the multiple and complex needs of PWUD (Rhodes, 2002). Although Rhodes’ framework explores how micro environments can shape individuals’ drug harms, our work documents how the framework can be utilized to explore the way environments shape service providers’ practices, including the challenges they face when providing services to PWUD.

Acknowledgements

We would like to thank all of the co-investigators and collaborators on this team for their helpful advice and guidance: D. Bailey, S. Campbell, J. Fraser, C. Jahn, F. Keough, A. Leard, C. MacIsaac, C. Porter, N. Scott, T. Warner, and D. Warren. As well, we extend our gratitude to members of the Advisory Committee for their support of this study: C. Davison, D. Ling, P. Melanson, M. O'shaughnessy and G. Rideout. A special thank you to the research co-ordinators A. D'sylva and D. LeVangie, as well as research assistants K. Adamson and J. Pasiciel.

We would further like to extend a sincere thank you to all of the study participants who so graciously gave of their time to participate in this study. We could not have conducted this study without their participation.

Declaration of interest: The authors report that there are no conflicts of interest. We would like to acknowledge and thank the Canadian Institutes of Health Research (CIHR) for funding this study through the HIV/AIDS Community-Based Research Program (Project #CBR – 104068).