Barriers and Facilitators to Harm Reduction for Opioid Use Disorder: A Qualitative Study of People With Lived Experience

Abstract

Study objective:

Although an increasing number of emergency departments (ED) offer opioid agonist treatment, naloxone, and other harm reduction measures, little is known about patient perspectives on harm reduction practices delivered in the ED. The objective of this study was to identify patient-focused barriers and facilitators to harm reduction strategies in the ED.

Methods:

We conducted semistructured interviews with a convenience sample of individuals in Massachusetts diagnosed with opioid use disorder. We developed an interview guide, and interviews were recorded, transcribed, and analyzed in an iterative process using reflexive thematic analysis. After initial interviews and coding, we triangulated the results among a focus group of 4 individuals with lived experience.

Results:

We interviewed 25 participants with opioid use disorder, 6 recruited from 1 ED and 19 recruited from opioid agonist treatment clinics. Key themes included accessibility of harm reduction supplies, lack of self-care resulting from withdrawal and hopelessness, the impact of stigma on the likelihood of using harm reduction practices, habit and knowledge, as well as the need for user-centered harm reduction interventions.

Conclusion:

In this study, people with lived experience discussed the characteristics and need for user-centered harm reduction strategies in the ED that centered on reducing stigma, treatment of withdrawal, and availability of harm reduction materials.

Editor’s Capsule Summary

What is already known on this topic

topicEmergency department (ED)-facilitated treatment for patients with opioid use disorder is important for reducing associatedmorbidity and mortality.

What question this study addressed

The authors conducted interviews with individuals with opioiduse disorder to identify barriers and facilitators to opioid use disorder-related harm reduction strategies.

What this study adds to our knowledge

Individual experience of stigma, housing instability, and inability to obtain naloxone are reportedbarriers to opioid use disorder treatment.

How this is relevant to clinical practice

Strategies to reduce stigma toward EDpatients with OUD and naloxone dispensing programs may improve harm reduction.

INTRODUCTION

Background

Emergency departments (EDs) play an important role in the treatment of patients diagnosed with opioid use disorder including opioid agonist treatments, such as buprenorphine and methadone and naloxone prescribing and distribution.^1–4 However, the morbidity and mortality from opioid use disorder extends beyond overdose deaths, as demonstrated by recent increases in injection-related infections, such as endocarditis, spinal epidural abscess, skin and soft tissue infections, HIV, and hepatitis C.^5,6 “Harm reduction” refers to evidence-based approaches that lessen the harmful consequences of drug use (eg, overdose deaths and infections) and was developed in response to “abstinence-only” approaches to treatment, and built on a respect for the rights of people who use drugs.⁷

Importance

The ED is often the first setting where patients with an injection-related infection and those with opioid use disorder receive medical care. Emergency clinicians, having seen the consequences of opioid misuse, are increasingly motivated to help patients prevent the negative outcomes of drug use.² As EDs are engaged in the treatment of opioid use disorder, opportunities exist for clinicians to facilitate harm reduction practices. In order to design effective, high-value harm reduction interventions for the ED, it is imperative to understand the perspectives of individuals who use drugs.

Goals of This Investigation

We sought to understand participant knowledge and use of harm reduction practices, identify barriers and facilitators to use of these practices, and explore how EDs could help at-risk patients use harm reduction resources.

METHODS

Study Design and Setting

We used qualitative methods to conduct semistructured interviews with a convenience sample of individuals with lived experience with illicit opioid use. This study was nested in a study of patient perspectives on starting medication for opioid use disorder in the ED. The second half of the interviews aimed to address knowledge of, barriers to, and facilitators of harm reduction and to solicit concrete suggestions regarding ED resources and clinician-patient interactions (Appendix E1, available at http://www.annemergmed.com)^.8 The research team developed an interview guide, piloted it, and revised the guide. The study is reported according to Consolidated Criteria for Reporting Qualitative Research standards for qualitative research.⁹ The local institutional review board determined the study exempt human subjects research.

Selection of Participants

Between May and June 2021, we recruited participants with opioid use disorder by convenience sampling through several means. We recruited patients from the Baystate Medical Center ED, a tertiary care academic center in Western Massachusetts with >11,500 patient visits/year. Patients presenting to the Baystate Medical Center ED with an opioid-related complication (overdose, withdrawal), history of an opioid use disorder diagnosis, or history of opioid agonist treatment use were identified from the tracking board by a research assistant. The study team member discussed the presence of inclusion and exclusion criteria with the treating clinician and approached patients who met inclusion criteria. Additionally, fliers were placed in the ED encouraging anyone with a history of an opioid use disorder diagnosis to volunteer for the study through a confidential online registry. To recruit patients from opioid agonist treatment clinics, we shared fliers with directors of 6 opioid agonist treatment clinics encouraging patient participation through self-enrollment. Patients met inclusion criteria if they were aged ≥18 years, able to speak conversational English, and had a history of opioid use disorder. Exclusion criteria included cognitive impairment/lack of capacity as determined by clinician and/or investigators, homicidal ideation/aggression, or in police custody.

Data Collection

Interviews, approximately 30 minutes long, took place in person or through video conferencing software or phone call. Prior to starting the interview recording, interviewers explained the study and obtained verbal consent using a standardized script. One to 3 members of the research team were present for each interview based on availability. Interviewers conducted participant checking (ensuring they understood participant’s points) in real time; transcripts were not returned to participants. Participants received a $20 gift card to a local or national store of their choice. Interviews were deidentified and transcribed verbatim.

Research Team and Reflexivity

Interviews were led by 1 to 3 team members: 1 woman emergency physician investigator with significant previous experience with qualitative methods (EMS) and 2 medical students (SAB [woman] and BP [man]) who received training in qualitative methods by experienced qualitative researchers, had at least 1 year of clinical experience providing care to individuals with opioid use disorder, and conducted a minimum of 2 proctored interviews. Coinvestigators were practicing physician-investigators, research staff, and an individual with living experience with opioid use as well as experience with both harm reduction practices in the community and research. The research team made it clear at numerous points in the interview that there were no right answers, that answers were completely confidential, and that participants could skip any questions they preferred not to answer.

Primary Data Analysis

Interviews were transcribed verbatim, and a research assistant reviewed the transcription for accuracy and readability. Qualitative data management software was used for coding and analysis (Dedoose Version 8.1, SocioCultural Research Consultants, LLC). We used reflexive thematic analysis, shifting between inductive and deductive modes.¹⁰ A 3-researcher team comprising 1 researcher with extensive experience with qualitative methods (LMW) and 2 research assistants read and reread the transcripts, making notes. Then, each reviewed the first transcript and created an initial coding framework using comparison and consensus. At this point, initial themes were felt to fit within the Fogg Behavioral Model (Figure 1) and characterized accordingly.¹¹ This model posits that an action (such as a harm reduction practice) occurs when there is enough motivation and enough ability, and if one is low, the other must be higher to compensate.¹¹ Viewing themes in the context of this model, each theme was characterized as increasing or decreasing motivation and/or increasing or decreasing ability. Each transcript was coded independently by 2 research team members using this coding framework. The study team discussed all codes and themes, and all transcripts were independently recoded by 2 members of the research team. The final codebook can be found in Appendix E1. The study team then generated initial themes from the data. Themes were reviewed and refined in the context of the full dataset and revised by discussion and consensus within the study team. We obtained adequate sample size through our analysis of information power, in accordance with recommendations for reflexive thematic analysis.^10,12

Figure 1.

Fogg Behavioral Model applied to harm reduction.

Triangulation

To explore themes in more depth and in another context, after coding was complete, the results were brought to a focus group of 4 individuals with lived experience with drug use for feedback. These participants were recruited through a community mutual-aid organization run by and for women who use drugs and engage in sex work. This group verbally consented and received $25 and lunch for their participation. Two members of the research team (a research assistant and a medical student) explained the goals and findings of the study, sought open-ended feedback on all themes as well as specific feedback on certain themes, and took notes. The notes were reviewed and discussed by the study team within the context of the themes to provide richer context and improve reflexivity.

RESULTS

Of the 37 people indicated potential interest from posted fliers, 19 were interviewed. Seven ED patients were approached and interviewed, of whom 6 completed the harm reduction portion. We met thematic saturation and ended enrollment prior to scheduling the remaining individuals who provided contact information through the fliers. Participant characteristics are displayed in Table 1. No participants had existing relationships with interviewers. Participants’ median age was 34 years (Table 1). Themes, subthemes, and representative quotations are displayed in Tables 2 to 4. Central themes included motivation and ability to engage in harm reduction strategies, mediated by several subthemes. Specific harm reduction strategies discussed are displayed in Figure 2.

Table 1.

Participant demographics.

Characteristic	Values (N=25)
Gender (self-identified), n (%)
Woman	15 (60.0)
Man	10 (40.0)
Nonbinary	0 (0)
Age (median, 34 y; IQR, 30 to 40 y), n (%)
18-30 y	7 (28.0)
31-50 y	15 (60.0)
>50 y	3 (12.0)
Race and ethnicity (self-described), n (%)
Black	1 (4.0)
Hispanic	2 (8.0)
Native American	1 (4.0)
White	20 (80.0)
Other*	1 (4.0)
Language spoken at home (other than English), n (%)
Spanish	4 (16.0)
ED visit in the past 5 y, n (%)
Yes	24 (96.0)
Experienced an opioid overdose,† n (%)
Yes	19 (76.0)
Insurance (only available for 21 participants), n (%)
Medicaid	17 (81.0)
Commercial	4 (16.0)
Highest level of education (available for n=25), n (%)
Eighth grade or less	1 (4.0)
Some high school	5 (20.0)
High school/GED	9 (36.0)
Some college or 2-y degree	8 (32.0)
4-y college	2 (8)
Opioid agonist therapy tried (all had tried at least 1), n (%)
Buprenorphine	22 (88.0)
Methadone	18 (72.0)
Both	17 (68.0)
Injection drug use, n (%)
Yes	16 (64.0)
No	9 (36.0)
Time since last unprescribed opioid (available for n=23), n (%)
<1 wk	5 (21.7)
1 wk to 6 mo	5 (21.7)
6 mo to 2 y	4 (17.4)
>2 y	9 (39.1)

IQR, Interquartile range.

^*Participant self-described as “other.”

^†Overdose was self-defined by participants and based on the description and was generally equated as requiring emergency care and/or naloxone.

Table 2.

Representative quotations of barriers and facilitators related to ability.^*

Accessibility
Need supplies on hand	“I wouldn’t have taken the time to do [go to syringe service program]. Because everything was about that money, get drugs, get high, like that was it. There was no any kind of extra steps that I wanted to be taking.” (15)
Proximity	“I have Narcan in my apartment, I have Narcan in every room. Always. I’ve saved people’s lives because I’ve had Narcan on me when nobody else did.” (16)
SSPs	“I would go to a needle exchange, so I could get clean syringes. Like tourniquets, rubbing alcohol, like wipes. We would get even condoms and all that kind of stuff. We’d get bleach for the needles if you had to reuse needles for some reason, because sometimes you don’t have access to it. And they show us how to clean it safely. So you’re not injecting bleach. And we’d also get Narcan.” (19)
Providing medication in hand more important than prescription	“They should just hand you the box [referring to naloxone]. Because a lot of addicts, if you give them a prescription, they’re not going to go in and fill it.” (24)
Knowledge
Knowledge of how infections work	“If I’m on a bender, and I see a little dirt. because I’m swimming in my own filth anyway I’m not going to die from that. I don’t know, I have a weird outlook when it comes to, you know, viruses, bacteria, and stuff like that. If you were to come like into my area, well that might be different. You know, if me and you ain’t equal…you got something a little bit different than my body’s had then I might get sick.” (8) “I lick the skin for it to be soft. I didn’t realize it was causing the bumps until someone told me.” (FG 3)
Habit-knowledge disconnect: culture of use	[Regarding licking needles] “Yep. I’ve done that. And especially when some of it starts to come out. And it’s bad for you, I know that but, you know, people don’t get what could happen by doing that…the desperateness of it is that much - you’ll risk whatever just to make sure that you don’t waste anything.” (12) “I would, you know, push the plunger up so that the liquid would get closer to the tip and then sometimes it would squirt out a little bit. So, like, instead of just putting it in my arm - I don’t know why I licked it –that’s actually a really good question. Because it’s disgusting. And like, you would think that somebody wouldn’t do that.” (16) [regarding water for a dissolving agent] “It’s kind of gross, but anywhere. I’ve gone from dirty bathroom sinks to toilets to puddles. Yeah, anywhere and everywhere.” (6)
Socioeconomic variables: houselessness limits accessibility	[regarding cleaning skin prior to injecting] “That it would be the least of anybody’s worries. Dude, I wasn’t even showering.” (3) [regarding water source] “Um, anywhere it could be - I’d either have a water bottle or be at a sink. You know, bathroom. Toilet water. A puddle, snow.” (19) [regarding water source] “It’s kind of gross, but anywhere. I’ve gone from dirty bathroom sinks to toilets to puddles. Yeah, anywhere and everywhere.” (6)
Stigma
Stigma in the community	“When it comes to Narcan, having people go to the pharmacy and having it on their med list is something that most people are just not going to want to do. So Narcan should be anonymously available to whoever wants it or needs it without having to have an extra step.” (11) “I used alone because I didn’t want people knowing that I was using.” (9)

FG, Focus group participant.

^*Major themes are in bold, and subthemes are in plain text. Participant number are in parentheses.

Table 4.

Representative quotations of need for user-centered strategies.^*

	Need for User-Centered Strategies
Fentanyl test strips	“That’s the whole idea, right? It’s the fentanyl that gets you high. You know what I’m saying? So, it’s like, why not? Compared to anything else I’ve ever done the highest I’ve ever got is off fentanyl.” (9) “I prefer the [bags] with fentanyl…I mean, everything’s fentanyl now. So, if it’s not fentanyl, then they usually kind of suck honestly.” (8)
Test doses	“I think that really anything having to do with like, testing a little bit out first or anything to know how potent the drug was work, for me, would have the reverse effect because I wanted my drugs to be as potent as they could be.” (15) “It’s like, when you’re finally, ‘Oh, I got my drugs to make myself numb’ you’re going to do as much as you can. Unfortunately, you don’t think about until after the fact when it’s too late, that you could overdose.” (20)
Naloxone use	[regarding a naloxone kit] “Talk about making it stressful in an already stressful situation. Somebody is overdosing and you’re sitting there trying to…put the thing together. terrible, terrible setup. But now they’re like the U-shaped ones that are ready to go.” (4)

^*Participant number are in parentheses.

Figure 2.

Harm reduction strategies.

I. Motivation (Increasing)

Knowledge and personal experience.

Personal knowledge and experiences with complications of injection drug use emerged as important motivators for some participants to engage in harm reduction, particularly in reference to infectious complications.

“Dirty blood sitting in a rig for a minute and you think is clean because you rinse them but you’re not a scientist so it’s not that clean. And then you got Hep C.” (18)

II. Motivation (Decreasing)

Lack of knowledge.

Participants had varying degrees of awareness of harm reduction practices. They unanimously identified that sharing needles was unsafe, and most reported attempts to avoid sharing needles. However, most participants revealed knowledge gaps in other forms of infection prevention. For example, many participants were not aware of the risk of soft tissue infections or endocarditis associated with injection drug use.

Lack of self-care: hopelessness and the overwhelming need to avoid withdrawal.

Most participants indicated some degree of lack of self-care and reported that this complicated engaging in harm reduction practices. Subthemes emerged that seemed to drive the lack of self-care, including the need to avoid withdrawal symptoms (above all else) and hopelessness. Participants described brutal withdrawal symptoms, and their fear and/or experience with these symptoms meant prioritizing the speed of drug use over safety at times.

“I was concerned about just getting sick. I didn’t want to feel like shit, puking you know…. That’s it. But, concerned about overdosing? No.” (8)

Many participants described hopelessness. Many participants denied a fear of overdose and alternatively endorsed either indifference or passive desire for overdose. Others also highlighted the fine line that many people were chasing between achieving an intense “high” and overdosing. Hopelessness was tied to socioeconomic factors, often lack of housing and poverty.

“In the grips of my addiction, it was like the lowest part of my life. And when you’re using every day and like doing all this nasty stuff just to get high so you’re not sick…you’re not living…. you kind of hope you die because you think it’s impossible to stop using. So, the overdose doesn’t scare you. If anything, you kind of like, glorify it, because it’s like the highest high that you’re chasing.” (19)

Stigma.

Participants discussed the power of stigma, the experience of negative and unfair beliefs associated with being someone who uses drugs, which contributed to difficulty obtaining harm reduction supplies, engaging with treatment, and seeking medical care. They described a pervasive sense of being judged, the assumption that symptoms were due to drug use/intoxication rather than a medical issue, and experiences in which they lacked autonomy as their drug use was perceived as suicidal by the treating team. Although participants felt that offering harm reduction supplies in the ED would be beneficial, some participants discussed a fear of the ED and health care as a barrier to accessing these resources.

However, participants noted stigma more widespread in the community, even among other individuals who use drugs, that made harm reduction more difficult. Specifically, participants described engaging in behaviors they knew were counterproductive to their health due to stigma, such as using drugs alone or avoiding obtaining new supplies due to a lack of anonymity.

Habit–knowledge disconnect.

Many participants explained that although they were aware of harm reduction practices, habits and/or rituals surrounding their drug use frequently limited use of those practices. For example, several participants reported licking their needle prior to injection despite knowing this was not safe, stating that the mouth “is full of bacteria.” Although reported reasons included minimization of waste that may leak out the end of the needle or wanting to taste the drug, often the reason for licking the needle was described as ritualistic, exposing the propagation of injection techniques between generations.

III. Ability (Increasing)

Accessibility.

Participants uniformly noted that provision of harm reduction supplies increased the likelihood that they would use the supplies. Although syringe service programs (SSPs) were identified as a critical source of harm reduction supplies, their accessibility and proximity were sometimes prohibitive. Participants noted it was important to meet people where they are physically, which could include the ED.

“I would use any needle. I didn’t care…that’s the problem…I know about the needle exchange, I would go there and I would still pick a needle up off the ground. Because that’s just how insidious the whole thing is, it’s like - going 20 minutes another way when it’s right here? It’s a lot.” (3)

IV. Ability (Decreasing)

Accessibility: socioeconomic determinants.

Additionally, socioeconomic determinants, such as housing, were reported as an important mediator to engagement with harm reduction. Lack of resources and housing made harm reduction techniques more difficult, particularly accessibility of water for preparation of opioids for intravenous use.

Lack of user-centered strategies.

When asked about some specific harm reduction supplies and strategies, participants highlighted a rift between the intent of the harm reduction strategy and their reality, making the task more difficult. For example, naloxone prescriptions from the ED and referrals to locations to obtain naloxone were insufficient for most of our participants. Handing out naloxone, on the other hand, mitigated the difficulty associated with filling a prescription and helped overcome low motivation and anticipated stigma. Additionally, participants questioned the value of fentanyl test strips as a harm reduction technique. Not only were very few participants aware of this tool, but when the concept was explained by the interviewer, most participants reported that knowing the heroin contained fentanyl would make the drug appear more enticing. Fentanyl was perceived as either ubiquitous or indicative of a better or adequate high, depending on opioid tolerance.

“I wanted the drugs that have fentanyl in them. I would read articles and be like, ‘This is something that someone OD’ed on, that’s what I want to use because that’s going to get me high.’” (15)

Similarly, few participants reported ever using test doses of opioids. Although participants acknowledged the rationale behind a test dose, most reported the urgency to avoid withdrawal generally limited use of test doses. Participants who used test doses reported using them in 2 specific circumstances: upon relapse or when obtaining drugs from an unfamiliar source.

V. Triangulation

The focus group comprised 4 women aged 23 to 60 years. Three reported illicit opioid use within the past 24 hours. Overall, the focus group affirmed the themes and added some depth and context, particularly the importance of convenience and availability and the threats to motivation.

Triangulation participants underscored that the fear and pain of opioid withdrawal is a critical barrier to harm reduction: the need to avoid withdrawal trumps the knowledge, motivation, and ability to engage in harm reduction strategies that individuals may otherwise use. The physical and psychological pressures of withdrawal meant that delays of seconds to minutes became too much. They noted that being on methadone mitigated this urgency, allowing more time for use of harm reduction techniques.

Triangulation participants provided context on potential implementation in the ED. For example, they noted persistent fear of the health care setting due to prior experiences with stigma, dehumanization, and loss of autonomy. They reiterated stories of health care providers being judgmental and how this makes them hesitant to seek needed care.

Reflecting on findings from the interviews, triangulation participants brought up additional context for themes. First, they reported that, due to stigma, many patients with opioid use disorder feel pressure to tell health care workers what they think they want to hear (eg, “I’m never going to use drugs again!”), and this may be a barrier to talking about harm reduction and acceptance of supplies. Second, participants reported that changing how people who use drugs are treated in hospitals, including addressing stigma and offering improved treatment of withdrawal, would help people who use drugs feel safer in health care settings, allowing for more productive conversations around harm reduction.

Lastly, triangulation participants clarified that the “culture of use” is not always a barrier to harm reduction. People who use drugs can learn effective practices from each other, and these networks can be harnessed to spread harm reduction strategies. This positivity did not extend to recovery coaches. Triangulation participants were wary of recovery coaches—they felt that recovery coaches’ focus was often singularly centered on abstinence, and this created a barrier to harm reduction conversations.

LIMITATIONS

We employed convenience sampling and may have oversampled individuals engaged in treatment for opioid use disorder, which could lead to selection bias and an overestimation of harm reduction practice knowledge. Alternatively, by recruiting participants from the ED, we may have oversampled people who use drugs and also have more tenuous social situations in terms of housing and income, as people who have more resources are less likely to get care in the ED. People with more resources may have different attitudes and practices around drug use and harm reduction (eg, the strong theme of hopelessness was likely a confluence of socioeconomic determinants). For feasibility reasons, our study team was composed of clinicians and medical students, and different study team members were present during interviews. Additionally, all study team members were White. Despite assurances of confidentiality, it is possible that participants may not have been completely comfortable disclosing their thoughts to the interviewers and study team members and introduced social desirability bias. Additionally, although many participants had lived in various places in the United States, the interviews took place in Massachusetts and may not reflect the perspectives of individuals in other regions. Further, we excluded non-English speaking patients.

DISCUSSION

In this study exploring perspectives on harm reduction related to opioid use disorder, participants described themes related to motivation and ability to engage in harm reduction, notably, availability and proximity of harm reduction supplies, knowledge of harm reduction practices, and barriers to self-care, such as painful withdrawal, stigma, and hopelessness.

The results of our study can be viewed in the context of the Fogg Behavioral Model.¹¹ The more difficult something is to do (ability), the higher the motivation must be for the action to take place. In contrast, easier actions require less motivation. Prompts or facilitators can fail to generate action when motivation and the ability to engage in the behavior are low but can succeed when 1 of the 2 factors is high (easy to do or high motivation). In our study, the proximity of harm reduction supplies was paramount and increased the ability to execute harm reduction practices. This was, in part, because of significant demotivators: stigma, the need to avoid withdrawal, and hopelessness.

We found that the need to avoid opioid withdrawal was critical to engaging in harm reduction—powerfully influencing both motivation and ability. Historically, many clinicians have embraced the mantra, “alcohol withdrawal kills, opioid withdrawal is uncomfortable.”² It is apparent that a gap exists between the clinician’s understanding and the patient’s experience of opioid withdrawal. For example, one description of opioid withdrawal demonstrates aspects not captured by scoring instruments: “Yes, withdrawal includes nausea and vomiting; sweating; gooseflesh; and the need to constantly ‘kick’ out your legs, but the real horror is feeling as though you cannot bear simply being for another second, every second.”¹³ Our participants described a fundamental need to avoid withdrawal that precluded the use of harm reduction strategies that prolonged their torture. In reality, the most common harm from opioid withdrawal is return to use of opioids without harm reduction strategies. Continued use without skin cleansing, new needles, clean water, self-care, using in stable environments, using in company, and using in proximity to individuals with naloxone makes the individual more vulnerable to skin and soft tissue infections, endocarditis, and overdose death. Our participants’ voices underscored this new paradigm: the urgency and pain of opioid withdrawal leads to unsafe drug use and higher risk of death; in this way, opioid withdrawal can kill. Emergency physicians can aggressively treat withdrawal and initiate opioid agonist treatment, which may reduce the urgency associated with withdrawal.

Stigma emerged as a powerful demotivator to harm reduction and barrier to harm reduction. For example, although SSPs were identified as a source of new supplies and a knowledge bank on complications, many individuals did not want anyone, including those at SSPs, to know or believe that they were using drugs. Similarly, participants universally discussed a fear of the ED and health care due to significant stigma. In fact, participants discussed avoidance of the ED, even for life-threatening emergencies. These feelings were rooted in widespread experience with inadequately treated withdrawal, resulting in confrontational interactions and leaving against medical advice. Thus, although the ED could be an important touchpoint for provision of harm reduction resources for individuals with opioid use disorder, major hurdles include patient receptiveness based on profound historical trauma as well as clinician receptiveness due to stigma surrounding opioid agonist treatment, particularly methadone.¹⁴ Although methadone reduces the urgency and amount of drug use, methadone use carries significant stigma from health care clinicians, and few emergency physicians are comfortable initiating methadone.^2,8,15 This highlights the critical importance of destigmatization and appropriate care of patients who use drugs in the health care setting as a harm reduction technique. Participant-generated ideas to destigmatize harm reduction conversations and general ED care are presented in Table 4.

High-value efforts to increase uptake of harm reduction strategies must focus on user-centered, stakeholder-derived strategies. For example, fentanyl test strips were designed to help individuals identify fentanyl, theoretically as a deterrent. However, study participants viewed fentanyl in their drug supply as inevitable and desirable, a sign of a more potent product. This may be a result of sampling from an area where fentanyl is ubiquitous in the illicit opioid supply; however, prior studies of fentanyl test strips have had mixed results.^16–18 This highlights a need for a menu of harm reduction supplies and understanding of local contexts.

Additionally, our participants reported that receiving naloxone directly in the ED was preferable to a prescription, which might cost money and necessitated travel to the pharmacy, another stigma-filled environment.^19–21 If we want more patients to have naloxone, our findings suggest we should provide it in the ED. This mirrors findings in prior studies—few patients fill prescriptions for naloxone using insurance.²² In contrast, the liberal use of low-barrier opioid agonist treatment, such as buprenorphine and methadone, through a harm reduction framework of addressing the immediate concerns of withdrawal without requiring long-term adherence may help to increase trust between patients and providers. Accessibility is paramount.

In this study, we found harm reduction strategies in the ED must literally and emotionally meet patients where they are and bridge gaps in motivation and ability. Necessary changes include decreasing stigma associated with health care, consistent delivery of compassionate care, timely treatment of withdrawal, and provision of on-site resources.

Table 3.

Representative quotations of barriers and facilitators related to motivation.

Theme*	Representative Quotation (Participant Identification Number)†
Need to avoid withdrawal	“When you’re really sick and you got bags and you have a rig in your pocket…you don’t want to, you know, spend 20 minutes to get there. But I think it’s something that people should just have on them.” (2) [Interviewer: Were you concerned about overdosing?] “No, nope, not No. Not at all. You’re just more concerned about not being sick. I mean, you pray that you don’t die.” (1)
Stigma
Stigma in the ED	“I felt very judged and like, to the point that I let it get so bad because I did not want to go to the hospital. I was too embarrassed. And at first, I was going to tell them that I slammed my hand in the car door. But by the time I got there my hand was so big that it weighed six and a half pounds. Because that’s how swollen it was from the infection…And by the time I was admitted that night, it was all the way up to my shoulder.” (16) (When asked what EDs could do better) “Just make sure that there’s not a stigma or, at the very least, as a provider that does feel like a certain type of way about it, not showing that to the patient… being able to really treat people fairly because that just drives people away from wanting to come and seek help when they need it…may be if they even need medical help from something.” (15) “Sometimes I tell them…I’m never going to use drugs again! Because that’s what they want to hear.” (FG 1)
Stigma in the community	“I was reusing them a lot…I was too embarrassed to go to [syringe service program].” (15) “People think because you have Narcan on you, you must be a junkie. So you’re like, Oh, man, people are gonna look at me messed up and stuff.” (12)
Previous complication of drug use	“I did get like a cellulitis infection one time on my hand like a year ago…so I try to use alcohol because that scared me.” (10)
Hopelessness	“I didn’t care either way [about death] when I was that far into addiction. And you’ve probably heard it before, but if I knew someone overdosed off bags, I wanted those bags because they were strong.” (6) “So, if I did happen to overdose and not come back, it wouldn’t be like the worst, worst thing. Which I’m sure is how it exists for most people that, you know, you’re using because they don’t like the way you feel when you’re sober.” (1)
Feeling of invincibility	[Interviewer: Do you ever worry about overdosing?] “Yeah, I’ve thought about it. But I also felt like oh, that can’t happen to me. Until then, it does.” (13) “He thinks he’s invincible or something… And he thinks he can’t die, because he’s been brought back so many times…It did something to me to have to see someone die in your bed and then get revived. It did a lot to me. But then it didn’t faze him.” (1)

FG, Focus group participant.

^*Major themes are in bold, and subthemes are in plain text.

^†Participant number are in parentheses.