Characterizing peer roles in an overdose crisis: preferences for peer workers in overdose response programs in emergency shelters

Abstract

Objectives:

A growing body of research points to increasing peer involvement in programs for people who use drugs, although this work has focused primarily on naloxone education and distribution. This study extends this work by examining the roles of peers in leading a novel overdose response program within emergency shelters.

Methods:

Semi-structured qualitative interviews were conducted with 24 people who use drugs, recruited from two emergency shelters, as well as ethnographic observation in these settings. Interviews were transcribed and analyzed thematically with attention to peer roles.

Results:

Four themes emerged from the data. First, participants discussed the development of peer support through relationship building and trust. Second, participants described a level of safety using drugs in front of peer workers due to their shared lived experience. Third, peer workers were described as favorable compared to non-peer staff because of nominal power dynamics and past negative experiences with non-peer staff. Last, given the context of the overdose crisis, peer worker roles were often routinized informally across the social networks of residents, which fostered a collective obligation to respond to overdoses.

Conclusions:

Findings indicate that participants regarded peer workers as providing a range of unique benefits. They emphasized the critical role of both social networks and informal roles in optimizing overdose responses. The scaling up of peer programming in distinct risk environments such as emergency shelters through both formal and informal roles has potential to help improve overdose prevention efforts, including in settings not well served by conventional public health programming.

1.0. Introduction

Many settings in North America are experiencing overdose epidemics due to the proliferation of opioids and other drugs adulterated with fentanyl and related analogs (Centers for Disease Control and Prevention, 2017; National Drug Early Warning System Coordinating Center, 2016). According to provisional data from the United States, opioid-related overdose contributed to over 67,000 deaths from September 2016 to August 2017, a 13.3% increase from the previous year (Ahmad, Rossen, Spencer, Warner, and Sutton, 2018). In Canada, the overdose crisis has been particularly severe in the province of British Columbia (BC), leading to the declaration of a public health emergency in 2016 when the province experienced approximately 1000 drug-related overdose deaths. In response, a range of overdose response interventions has been implemented in various settings, including witness injection rooms and overdose prevention sites.

The involvement of people who use drugs (PWUD) has been a feature of many newly implemented interventions throughout Canada including in Vancouver, BC (Greer et al., 2016). Often referred to as “peers,” the leadership and involvement of PWUD in harm reduction services has been demonstrated to have numerous benefits. Previous qualitative research on take-home naloxone programs has focused on the importance of these programs in training PWUD how to use naloxone and in fostering connections to the broader community (Sherman et al., 2008), as well as other intrinsic benefits such as confidence, and empowerment (Banjo et al., 2014; Marshall, Perreault, Archambault, and Milton, 2017; Mitchell et al., 2017; Wagner et al., 2014). Qualitative research in this area has paid scant attention to the potential broader benefits of peer-support that have been shown to be valuable for other aspects of harm reduction (Hayashi, Wood, Wiebe, Qi, and Kerr, 2010; Kerr et al., 2006; Small et al., 2012; Weeks et al., 2009) as well as clients’ perspectives of peer roles in naloxone programming. One study in China on peer-led overdose response outreach teams found that clients had a strong preference for peer workers over emergency medical professionals based on trust issues and past negative experiences. Although it also found that participants did not want to always rely on peer workers to access naloxone and preferred to have immediate personal access to it (Bartlett, Xin, Zhang, and Huang, 2011). Herein, we seek to build on the existing literature by providing an analysis of the roles of peers in leading a naloxone overdose response program within emergency shelters in Vancouver, BC.

Two seasonal emergency shelters implemented “peer witness injection programs” whereby residents can use drugs in front of peer staff who play a supportive role and are able to respond to overdoses by alerting emergency medical services and administering naloxone and CPR. These shelters have designated “injection rooms” equipped with tables, chairs, and harm reduction equipment. While these services are mainly used by shelter residents, other PWUD in these neighbourhoods are allowed to use these services during daytime hours. In this study, we examined shelter residents’ experiences using drugs in an emergency shelter with a focus on their perceptions of peer workers.

2.0. Methods

Data collection occurred between March and April 2017. Twenty-four qualitative semi-structured interviews were conducted with PWUD at two shelters (see Table 1 for sample characteristics). Participants were recruited via word-of-mouth and were referred to our research team by shelter staff. Interview participants provided informed consent and each received $30 (CAD) honoraria for their time. Participants chose their own pseudonyms. A guide was used to facilitate the interviews. Interviews were audio-recorded, transcribed verbatim, and checked for accuracy by the lead author. Ethnographic fieldwork (30 hours) was also employed to witness overdose responses and related social interactions and used to enhance the trustworthiness of interview data (Creswell and Miller, 2000). Ethical approval for this study was obtained through the University of British Columbia / Providence Health Care Research Ethics Board. We employed a team approach in developing the coding framework for the data. Members of our research team each reviewed a portion of the transcripts individually and then met as a group and developed a coding framework using a priori themes as well as thematic categories that emerged from the dataset (Corbin and Strauss, 2015). NVivo 11 was used to manage and code the data into themes and subthemes for analysis.

Table 1.

Sample Characteristics (n=24)

Age
Average (mean)	41
Range	26–59
Gender
Women	6
Men	18
Ethnicity
White	13
Racialized/Indigenous	11
Emergency Shelter Sites
Site #1	12
Site #2	12

3.0. Results

3.1. Support

Participants reported that the support provided by peer staff extended beyond injecting in the witness injection room, and into other spaces as well as before and after participants were using drugs. While peer workers were compensated for providing support, participants characterized peer support as encompassing traits such as “trust” and “genuine caring.” Participants discussed the relationship-building that occurs between a peer worker and a shelter resident, where peer workers “get to know you on a personal level” (Kegan, Age 29). Support was described as comforting, relational, and involving trust: “It gives the addict a sense of comfort, right? It gives the addict a sense of okay, here is somebody who is willing to sit with me and relate with me” (Kevin, Age 37). Through the support provided by the peer workers, participants described how non-judgmental support accommodated their drug use.

3.2. Safety

Building on the sense of support that participants expressed in relation to peer workers, there was also a sense of safety felt by interviewees. This was discussed not just in terms of peer workers being in close proximity to participants when they were using drugs, but also due to elements of trust. Their presence made participants feel as though they would be safe in the event of an overdose. Safety was described through the peer workers’ actions as illustrated in the following quote:

You feel safer around them... they’re sitting there right beside you. And it’s like, ‘Why are you here? ‘ And they’re like, ‘I’m making sure you’re okay.’ And even just saying that it just sends a shiver down my spine. They’re there when no one else is. (Kegan, Age 29)

The ways in which safety was described was more nuanced than simply peer workers physically observing participants. It was also connected to a lack of power differences, social safety vis-à-vis shared experience, and perceived caring in contrast to their everyday experiences.

3.3. Preference for Peer Workers

A comparison between the job descriptions of peer workers and support workers shows close similarities in terms of roles and responsibilities. Despite these similarities, interview participants described their perceptions of, and engagement with, peer workers as different and more favorable compared to non-peer staff due to commonalities and a shared lived experience of using drugs and experiencing homelessness. As described by one participant: “it’s nice to have someone that understands your point of view and doesn’t look down on you” (Derek, Age 50). In terms of power dynamics, participants described past negative experiences with non-peer staff as including intimidation and an uneven power dynamic compared to peers. For example: “staff is just that authoritative person, right? Generally, people in our world get judged” (Keri, Age 37). Participants described a lack of judgement from the peer workers and perceived them as different based primarily on their lived experience.

3.4. Peer Role Routinization

While participants discussed the importance of the peer workers in the shelter, this role was routinized across the shelter and taken on informally by shelter residents. Given that peer workers worked a limited number of hours at the shelters (on average 6 hours per evening), and against the backdrop of an overdose crisis, participants discussed the crucial importance of taking on some of the responsibilities of the peer worker role as a form of collective responsibility. This was done informally to ensure that other residents would still feel safe and have support in the event of an overdose. This is not to say that support staff did not play a role in providing harm reduction education and overdose responses; rather, given the small number of support staff per resident (2 per shift, supporting approximately 40 residents/site) and the frequency of drug use across the sites, participants felt an obligation to fill this role. Participants discussed doing checks on others using non-designated spaces: “I do regular checks now because of people overdosing in the bathrooms so many times” (Smoke, Age 26). Participants also let others know when they were using drugs to ensure someone was available to supervise them. For example: “I try to seek out as many people as I can [and] let them know that I’m going to be shooting up ...so people around me know they can look over ...and make sure I’m okay” (Kevin, Age 37).

4.0. Conclusions

In summary, our results suggest that participants regarded peer workers as offering a range of benefits. These were based on a shared lived experience and nominal power dynamics compared to support staff. Further, these findings illustrate how a sense of collective responsibility among residents was fostered and led to informal peer overdose responses.

Past qualitative research on peer-involved overdose response programs tends to focus on the benefits of take-home naloxone programs and the perspectives of PWUD on the training and administration of naloxone (Banjo et al., 2014; Mitchell et al., 2017; Wagner et al., 2014). Our study builds on this research by investigating the experiences of those on the receiving end of peer-led overdose responses. Similar to research undertaken in China where clients described a strong preference for peer workers over emergency medical professionals (Bartlett et al., 2011), participants in our study also favored the support of peers over non-peer staff, given past negative experiences with the staff at shelters. Further, while contextually different, our study similarly demonstrates the desire for PWUD to also respond to overdoses, as was evident in the routinization of the peer witness role.

It is apparent from our study that the involvement of peers as part of the witness injection program helped to optimize this service. These findings illustrate the need to expand peer roles in the delivery of shelter-based harm reduction programs, especially given the typically low staff-to resident ratio in such settings. This could include a collaborative approach, matching peer workers with other non-peer staff to potentially influence trusting relationships between PWUD and staff, particularly those that have historically discriminated against PWUD (e.g., medical professionals) (Dechman, 2015). Additionally, the routinization of informal peer support roles across shelter residents exemplifies a social phenomenon that warrants further investigation in drug use spaces, particularly during an overdose crisis. Quantitative studies have examined social networks of PWUD and how these inform the contexts of HIV risk (Latkin et al., 1995) and create effective HIV prevention linkages (Valente, Gallaher, and Mouttapa, 2004; Margaret R. Weeks, Clair, Borgatti, Radda, and Schensul, 2002). Within the context of an overdose crisis, our study highlights the ways in which shelter residents informally create such linkages. Further research should continue to examine these social networks and find ways to expand overdose prevention and naloxone programming across overdose risk environments.

Our study has its limitations. Our sample involved PWUD across two shelters and may not be applicable to other PWUD nor other drug use settings. Further, by focusing exclusively on the peer witness injection program, we may have overlooked the experiences of PWUD in engaging with staff and peers in other capacities (e.g., housing support). In conclusion, our study demonstrates the importance of peers in providing harm reduction support and overdose response in emergency shelters. Greater involvement of peers in harm reduction programming through both formal and informal roles can create more accessible and effective overdose response programming.

Highlights.

• Novel overdose response interventions are needed in emergency shelters.

• Peer-led overdose response programs offer a variety of benefits.

• Relationships with peer workers were built on a shared lived experience.

• Informal peer support roles optimized overdose response efforts.