“I don’t go overboard”: Perceptions of overdose risk and risk reduction strategies among people who use drugs in Rhode Island

Abstract

North America is experiencing an unprecedented overdose epidemic, with data estimating almost 110,000 overdose deaths occurring in 2022 in the United States (US). To address fatal overdoses in the US, community organizations and local health departments in some jurisdictions have expanded community distribution of naloxone, overdose prevention education, and other harm reduction supplies and services (e.g., fentanyl test strips, drug checking programs) to reduce harm for people who use drugs (PWUD). Understanding how PWUD manage overdose risk within the context of these expanded services is important for ensuring public health services are meeting their needs. Semi-structured qualitative interviews were conducted with 25 PWUD who were accessing harm reduction services in Rhode Island. Data were imported into NVivo where they were coded and analyzed thematically. Our findings demonstrate the complexity of managing overdose risk in the context of a fentanyl drug supply. While most participants were concerned about overdosing, they sought to manage overdose risk through their own harm reduction practices (e.g., testing their drugs, going slow) and drug purchasing dynamics, even when using alone. Study findings point to the need to implement and scale-up community-level interventions (e.g., naloxone access in public spaces, overdose prevention centers, community-based drug checking services) to better support PWUD within the context of the current US overdose crisis.

INTRODUCTION

North America is continuing to experience an unprecedented overdose epidemic, with almost 110,000 overdose deaths estimated to have occurred in 2022 in the United States (US) [1]. This epidemic has been driven by synthetic opioids, with provisional estimates suggesting that synthetic opioids were involved in approximately 65% of overdose deaths in the US from 2021–2022 [2]. In recent years, however, the overdose crisis has seen an increase in polysubstance-involved fatal overdoses, including a growing proportion of overdose deaths involving psychostimulants (e.g., cocaine, crystal methamphetamine) alongside opioids [3–4].

To address the increase in fatal overdose in the US, organizations and local health departments in some jurisdictions have increased overdose prevention education and expanded the distribution of naloxone and other harm reduction supplies (e.g., fentanyl test strips, safer smoking kits) and services (e.g., drug checking programs) to reduce adverse harm for people who use drugs (PWUD) [5–9]. While such harm reduction approaches have been documented to increase awareness about the unregulated drug supply and reduce overdose risk [10–12], research has highlighted how social-structural inequities (e.g., structural racism, gender-based violence, poverty) can complicate overdose risk management among structurally vulnerable PWUD and disproportionately impact sub-populations of PWUD (e.g., racialized and minoritized populations, unstably housed populations) [13–16]. Understanding the factors that impact overdose risk management and how overdose risk is negotiated by PWUD is critical to effectively addressing the needs of PWUD.

We sought to explore the perspectives of PWUD in Rhode Island on the overdose crisis and how people managed overdose risk. Rhode Island has remained significantly impacted by the overdose crisis, having the 12^th highest fatal overdose rate in the country in 2020 [17]. The impacts of the COVID-19 pandemic have further driven fatal overdose rates in the state, resulting in an approximate 25% increase from 2019 to 2020 and a 12% increase from 2020 to 2021 [18]. The local drug supply has continued to be complicated by fentanyl—and more recently, xylazine [19]—with approximately 75% of fatal overdoses in 2022 involving fentanyl [20]. A range of harm reduction services and overdose prevention interventions have been implemented in the state to address the overdose crisis, including the widespread distribution of naloxone and fentanyl test strips, as well as targeted harm reduction outreach in overdose hotspots. Given these ongoing efforts across the state, this study aimed to understand PWUD’s overdose risk reduction practices and engagement with harm reduction services.

METHODS

This study draws on qualitative data collected with 25 PWUD accessing harm reduction services in Rhode Island. Semi-structured interviews were conducted from August to September 2021 and sought to understand how PWUD were engaging with harm reduction services and resources, and implementing harm reduction strategies in their use practices. This qualitative study is part of a larger Rhode Island Harm Reduction Surveillance System (HRSS) conducted in partnership with the Rhode Island Department of Health and the Miriam Hospital to inform prevention and outreach activities in the state [21]. All study activities were approved by the Institutional Review Board at Lifespan Hospital System.

Participants were recruited through field-based recruitment methods at harm reduction partner organizations who provide services across the state. Most participants were recruited from sites serving Rhode Island’s urban core, with the remainder recruited from two cities outside of the metro-Providence area. Potential participants were recruited in-person, screened for eligibility on-site, and interviewed in a private room at the offices of the partner organizations. Participants were eligible if they were: 18 years of age or older, currently living in Rhode Island, and self-reported substance use other than marijuana or alcohol in the 30 days prior to their interview.

Interviews were conducted by three interviewers trained in qualitative research methods. Verbal informed consent was obtained from all participants prior to conducting interviews. Interviews were facilitated using an interview guide that included topics such as: ancillary service utilization, drug use patterns, drug purchasing dynamics (e.g., relationship with sellers), drug-related education (e.g., safer use practices, fentanyl education), overdose response (e.g., access to naloxone, naloxone administration), and gaps related to harm reduction. Interviews lasted approximately 30–60 minutes, were audio recorded, and transcribed by a professional transcription company and reviewed for accuracy by study team members. Participants received $45 cash reimbursement for their time. Harm reduction supplies (e.g., fentanyl test strips, naloxone) and other educational materials and supports (e.g., safer use practices, meal program locations) were available for access from the organizations where participants were recruited.

Data were imported into NVivo, a qualitative data management and analysis software where they were coded and analyzed thematically [22]. A preliminary coding framework was developed based on topics from the interview guide (e.g., service utilization, consumption patterns, overdose experiences, purchasing dynamics). The coding framework was refined through line-by-line coding in which emerging topics from the transcripts were added to the coding framework (e.g., trusting seller). All interviews were double-coded. Coders met regularly during the coding process to discuss any discrepancies in the application of codes, and discrepancies were resolved using a group consensus process. The coding framework was revised as new categories emerged during analysis, and transcripts were re-coded once final categories in the coding framework were established [22]. Following final agreement on all coding decisions, NVivo files were merged creating a central project file. The research team met regularly during analysis to review data and discuss emerging themes. This analysis focused perceptions of overdose risk, seller dynamics, and overdose risk reduction practices.

RESULTS

Participants averaged 38 years of age (see Table 1). All participants were cisgender, and most were men (72%), with the remainder women (28%). Most participants were white (68%), with the remainder being Hispanic (20%), Indigenous (8%), and multi-racial (4%). Over half (56%) of participants were unhoused. Of the 25 study participants, 76% used drugs daily, while the remainder used drugs approximately two to four times per week. About half (52%) of participants had experienced one or more overdose events (range: 1–8+), 32% had never overdosed, and the remainder (16%) were unsure or did not respond.

Table 1.

Participant demographics (n=25)

Participant characteristic	n (%)
Age  Mean	38.7 (range: 25–59)
Race & ethnicity  White  Indigenous  Hispanic  More than one race	16 (64%) <5 <5 <5
Gender  Woman  Man	7 (28%) 18 (72%)
Housing status  Housed  Unhoused2	11 (44%) 14 (56%)
Substances used in last 30 days1  Fentanyl/heroin  Crack cocaine  Cocaine  Crystal methamphetamine  Alcohol  Benzodiazepines  Marijuana  Other (e.g., Adderall, ecstasy)	19 (76%) 10 (40%) 9 (36%) 9 (36%) 8 (32%) 7 (28%) 15 (60%) 7 (28%)
Frequency of substance use  3–4 times per week  Daily  No response	<5 19 (76%) <5
Overdose history (last 12 months)  None  One  Two  Three or more  Unsure/No response	8 (32%) <5 <5 8 (32%) <5

¹Responses are not mutually exclusive.

²Includes couch-surfing, unsheltered, and staying in hotels/motels or shelters.

The impact of fear on drug use practices

Participant narratives underscored how fear was a prominent factor shaping their drug use practices and overdose risk mitigation approaches. Most participants’ fear was centered around the constant unknowns related to the local drug supply and the subsequent risk of overdose (i.e., getting a “bad batch”). One participant explained:

[I’m worried] if I’m going to get a bad batch and then I might die. That’s the fear. I don’t want that to happen. Like there’s a lot of people dying from it so you don’t know what could be - anything [could be] in it. [‘Martin,’ 38-year-old white man]

Other participants echoed similar sentiments, stressing that you “just never know” when an overdose might happen. However, for many participants, there was a need to manage withdrawal symptoms which led to ongoing uncertainty about their wellbeing. ‘Blake,’ a 38-year-old white man explained:

It’s never knowing what you got. Even though I try being safe with those little five shots, I get that fear of ‘can this one be it?’ I know I have no choice cause I’m sick and I have to use, but that fear is still in me like, ‘Oh man, what if this is the one that might put me out cause all that’s going around?’ Like that shit that’s stronger than fentanyl now…it’s just crazy. You just never know. …I’m sure a lot of us users out there fear that.

While most participants described being afraid of having an overdose, others underscored how they had more recently become concerned about overdose risk after people in their networks had overdosed. Among these participants, the increased visibility of overdose led them to be more concerned about their own overdose risk. One participant shared:

It’s just a scary thing to see these people dying and overdosing off this stuff. I don’t want to die from fentanyl. It’s almost like they put that out there just to kill people. [...] I never really asked for it [fentanyl test strips]. I don’t think it was that serious, but the more these people keep dying, it’s starting to be serious. I need to really be careful. [‘Jamie,’ 39-year-old Indigenous man]

For others, perception of overdose risk was shaped by a lack of prior overdose events, and their preferred substance (e.g., stimulant vs. opioid) and consumption approach (e.g., inhalation). One participant who predominately smoked crack explained:

As long as you’re not sticking a needle in your arm with the drug, I don’t fear nothing from the pipe. …Every last piss test I took in the hospital was dirty for fentanyl. [Are you worried about that?] Nah. I don’t know why it’s in it [crack], but it gives you a better buzz to tell you the truth. It takes the edge off the coke, you know what I mean? [‘Rodney,’ 58-year-old white man]

Similarly, ‘Gabriel,’ a 36-year-old Hispanic man described: “I like getting high so that’s why I do it…I haven’t overdose either so I don’t know what that feeling is. So I’m not really scared of it to be honest.”

Every day risk reduction strategies during use

Most participants readily acknowledged the risk of overdose given the proliferation of fentanyl in the unregulated drug supply, which underscored the need to engage in harm reduction practices to reduce potential harm. One participant who injected fentanyl explained:

I try to be very careful, you know. Like I just try to be careful when I get up and I sniff a few bumps first. And I wait, you know what I mean. I don’t go overboard. Cause I don’t want to die. It’s scary. That stuff is real serious. [‘Crystal,’ 44-year-old white woman]

Similarly, participants described how reducing the amounts they used and using little bits at a time were important practices to engage in to reduce the overdose risk. ‘Melissa,’ a 59-year-old Hispanic woman, shared:

When you’re in this life you don’t know what’s going to happen, you know? It could happen to anybody if we don’t be careful…I haven’t gotten an overdose…I’ve been lucky. So I don’t be greedy. […] Sometimes you can taste it – like when the heroin kind of numbs your mouth, your tongue. And crack – it do the same thing. Most of the time it’s better to taste it. Don’t put something in your body without [trying] it.

Although participants acknowledged the risk of overdose, many preferred to use alone due to needing to address withdrawal symptoms and to increase pleasure. In these instances, participants described various approaches they used to minimize overdose risk when using alone. ‘Marcus,’ a 38-year-old white man, explained how he modified his consumption practices:

Every time when I use, I use by myself. [...] Cause I’ve heard bout a lot who do that [fentanyl] overdosed on it. So yeah, it scares me sometimes, thinking about overdosing. Cause like I said, I got a 15-year-old daughter I gotta think about. So yeah. Sometimes like I get worried about it. [...] [But] I just use a little bit at a time.

Others who preferred using alone would seek out semi-public places (e.g., public washroom) to increase chances of someone being able to respond in the event of an overdose. ‘Ariana,’ a 26-year-old white woman, described:

I went in a [public] bathroom and I snorted it cause I was sick. …[I’m worried] if I’m gonna die…or if it’s bad, you know what I mean? Or if I go out, you know, overdose, and no one’s around or no one has Narcan or all that bullshit, you know? Like, what if no one has Narcan? You die.

Drug purchasing dynamics

Reducing risk through purchasing dynamics

While most participants relied on numerous sellers, many described how their relationships with sellers were defined by longer-term engagement, trust in the product, and communication which were seen as critical to reducing overdose risk. One participant explained: “I spend a lot of money so they don’t fuck with me. And I don’t buy $20 pieces, I buy hundreds at least” (‘Rodney,’ 58-year-old white man). While some participants described purchasing from only a few sellers—usually a “primary” seller and an “option B”—they simultaneously underscored the challenges of their seller relationships. ‘Jessica,’ a 29-year-old Indigenous woman who only purchased from one seller explained:

It’s kinda annoying. You go to his house he takes 30–45 minutes to fricken come downstairs. Then it’s always an excuse why it took so long. The bags are half the size they’re supposed to be. But I’ve been dealing with him for so long I don’t wanna go to somebody else and either get burn, get garbage, whatever and you know, get something I don’t know what it is.

Like ‘Jessica,’ other participants described weighing the difficulties they experienced at times (e.g., tardiness, smaller amounts) with risks involved of finding a new seller. One participant explained:

I usually just go through one person because I know what it is and trust what it is. […] If I can’t [find them] I usually don’t [use] because it fucks your health if it’s bad. …I’ve seen people get like bath salts and they just bought shit that they thought was dope or heroin of somebody and they were just like selling them this shit. […] So I’ve been going to the same person for less than a year so, not that there’s some kind of like magical number you know. How well can you really trust somebody anymore, you know? [‘Jodi,’ 39-year-old white woman]

In doing so, participants underscored the risks involved with drug purchases and how maintaining a few seller relationships was a way to minimize overdose risk.

For others, the quality of the product served as a catalyst for trusting their sellers. ‘Travis,’ a 40-year-old multi-racial man explained:

I have three main people, nobody else. Because I know I have never gotten anything bad, anything waste, or anything that I didn’t order or want. It’s always quality, quantity, and supply and demand.

Many participants described how communication and what they viewed as consistency in the product contributed to them trusting their seller. ‘Robert,’ a 46-year-old white man explained:

I have like three [sellers]. Two for crack and I only use one for fentanyl – the same one all the time, so I know what I’m getting. Cause he tells me how he mixes it so I’m comfortable with it. …He hasn’t sold to anyone that’s overdosed yet I think. And he does have a lot of customers he said. And he goes, ‘I honestly haven’t heard of none of my customers dying.’

Like ‘Robert,’ other participants described how minimizing the number of sellers they engaged with was imperative to reducing risk of overdose. In these instances, participants had either experienced an overdose after purchasing from different sellers or had heard of people having adverse outcomes after purchasing from new people. ‘Diego,’ a 25-year-old Hispanic man who described having four “go tos” shared:

I have two primary ones, one for the opiate and one for the uppers. You can’t trust nobody nowadays…but for that [buying] I do. Cause every time that I’ve went to somebody new…like you get burned or they give you something that makes you overdose. So I just try to stay with the same people because I know what I’m getting, you know where it’s coming from. Like I never had an issue or if it’s something new they tell me.

Social-structural impacts on purchasing dynamics

Participant narratives underscored how trust as it related to purchasing dynamics was often complicated by the changing supply. As a result, participants often described the tension between trusting their sellers’ product while also acknowledging that the supply was changing. One participant, ‘Mario,’ a 32-year-old Hispanic man, shared how he still used fentanyl test strips even though he purchased what he described to be heroin:

[I use] heroin, no fentanyl. …When I do get it I only got one person that got the real stuff. But it’s real rare to get it around [here]. But he goes all the way to [city] to get it. [Interviewer: Do you still test it with a test strip?] Oh sure, always cause I can’t do fetty. My mom died because of fetty – I’m not gonna die.

Utilizing harm reduction strategies, like fentanyl testing strips, was characterized as important for further reducing overdose risk within the current drug supply.

For other participants, their ability to trust sellers and limit the number of people they purchased from was impacted by the criminal-legal system in ways that increased their risk of overdose. One participant explained the challenges of trusting sellers:

Everybody out there, they’ll be taking your money or they’re goin to take something out of the bag or they’re going to rob you. […] When I got out of jail I lost all my connections. …It’s kind of hard to find one main guy. So I’ve been going to different ones… like 10 or almost 15. … I can’t get a hold of them or I don’t know they’re not on anymore so I ask somebody to get it for me. And it’s no good and then I ask another person to get it. So it’s all different people you know what I’m saying? [‘Martin,’ 38-year-old white man]

The shifting of purchasing dynamics based on interactions with the criminal-legal system was echoed by other participants who underscored how tenuous some relationships with sellers could be:

They’re always different places, different people. It’s not the same dealer. Some people come out of jail and start selling and they start using again and then that dealer don’t sell no more - cause he started using again and he becomes a customer, instead of the dealer. […] Sometimes you do [have a preferred seller] but then it changes if something happens. If he’s not a user, he gets busted. [‘Melissa,’ 59-year-old Hispanic woman]

Managing withdrawal and substance use needs

Although participants shared concerns about overdose risk, only three participants reported having purchased from one seller in the month prior to their interview. For many participants, purchasing from multiple sellers was largely shaped by their polysubstance preferences, availability of sellers and product, and an effort to minimize wait times of transactions. One participant explained how timeliness often impacted who they relied on as their primary seller: “They [primary seller] show up…they’re fast. But they’re, I don’t know, they’re alright for the time. They’re alright” (‘Omar,’ 51-year-old Hispanic man). While many participants who utilized multiple sellers described the “first come first served” nature of their purchasing dynamics—in that they purchased from the seller that was quickest to arrive—such practices were often driven by the need to manage their withdrawal symptoms. One participant who had experienced multiple overdoses described:

I thought it was straight H [heroin], but it wasn’t it was mixed. Let’s call it the fentanyl – yeah. And screwed me up. I got it off a person that wasn’t my normal guy, you know what I mean? But when you’re desperate, you’ll get it, you know what I mean? [‘Rodney,’ 58-year-old white man]

For many participants the ongoing negotiation of balancing overdose risk with withdrawal management often undermined their preferred drug use purchasing dynamics. ‘Crystal,’ a 44-year-old white woman, shared how her efforts to reduce overdose risk by purchasing from a primary seller were often complicated by her opioid withdrawal experiences:

I have my usual guy but then I seen somebody else so maybe like three [sellers]? …Because if my regular guy’s not around, he don’t have it, and I’m saying I have to get somebody else, you know? But otherwise I stick with my regular dude you know? It’s safer for me – the prices are good, he takes care of things. I’m so fucking scared of even doing that shit and I still do it. But when you’re sick it’s horrendous.

DISCUSSION

Our findings demonstrate the complexity of managing overdose risk in the context of a fentanyl-driven drug supply. While most participants expressed concern about an uncertain drug supply increasing overdose risk, participants simultaneously described how they sought to manage overdose risk through their own harm reduction practices (e.g., testing their drugs, going slow) and drug purchasing dynamics, even when using alone. Although most participants purchased from numerous sellers, they underscored navigating potential overdose risk by attempting to limit their purchasing networks. This was positioned as imperative for reducing potential harm within the context of their structural vulnerability.

Most participants described the potential of overdose within the current drug supply. However, for participants who predominately used stimulants or smoked substances, overdose-related concerns were not always expressed. Previous research has shown the need to expand harm reduction services and overdose messaging to be inclusive of substance type and methods of consumption given the proliferation of fentanyl across drug supplies [23–25]. Our findings further underscore the need for increased messaging about stimulants and harm reduction practices across substance type to better support people who consume a range of substances.

Importantly, participants described engaging in numerous harm reductions strategies when using (e.g., using fentanyl testing strips, reducing amounts used) and purchasing (e.g., building trust with sellers) drugs. While many of the harm reduction practices participants engaged in are in line with broader harm reduction messaging of overdose risk reduction, participants regularly described how they navigated overdose risk in ways that extended to seller relationships. Similar strategies, including the importance of social dynamics with sellers, have been documented in previous research as effective risk reduction strategies for PWUD [8, 11, 26]. However, recent research has demonstrated the adverse impacts of drug supply disruptions on overdose events, including through supply-side interdiction efforts (e.g., drug seizures) [27]. Our findings echo the importance of purchasing dynamics while also acknowledging how the social and structural inequities faced by structurally vulnerable PWUD (e.g., criminalization, housing instability) can complicate harm reduction practices. Given the importance of trusted seller relationships in harm reduction practices, there is a need to shift drug policy away from supply-side efforts which can reinforce overdose risk to evidence-based harm reduction efforts and decriminalization to better support PWUD. Implementing harm reduction supports and programs that strengthen the social practices PWUD engage in to reduce overdose risk is important for being responsive to the lived realities of structurally vulnerable PWUD.

Although participants in our study described engaging in harm reduction strategies, many still preferred to use drugs alone. Research has documented drivers of using drugs alone, including increased pleasure and privacy, and reducing risk of social harm (e.g., violence) [28–30]. Our findings echo this prior research and also underscore additional harm reduction practices participants engaged in even when using alone, including using in semi-public spaces (e.g., public washrooms). In shifting to more semi-public locations, participants sought to increase the opportunity of someone being able to respond in the event of an overdose. As more than half of our participants were unhoused at time of interview, use in public spaces was common. Expanding community access to naloxone within public and semi-public spaces (e.g., parks, parking garages, bus terminals) is imperative to reduce risk of fatal overdoses across settings. Additionally, implementing and expanding a range of overdose prevention interventions such as community spotting [31] and overdose prevention centers [32–33] are also necessary to reducing fatal overdose risk among people who prefer to use alone.

Our findings indicate a need to expand access to supply testing options for people who use drugs to further reduce overdose risk. This includes increasing access to personal testing approaches (e.g., fentanyl testing strips, benzodiazepine testing strips), as well as community-based drug checking approaches. Previous research has documented to acceptability and utility of these services at reducing overdose risk among people who use drugs [32–35], and the benefits of expanding drug testing services for sellers [36–37]. Legalizing supply testing options, including fentanyl test strips and community-based drug checking services, across US jurisdictions [38] and increasing funding supports for these programs is critically needed to better support the needs of PWUD at overdose prevention.

There are several limitations that should be noted. Participants were recruited from harm reduction services and may thus have received more extensive harm reduction and drug supply education than people less engaged with harm reduction and outreach services. As most harm reduction services in our setting are located in urban settings, future research should focus on rural areas to identify overdose risk management and needs among people in these settings. Further, all participants were cisgender thus obfuscating the experiences of non-binary and transgender people. White participants and men were also overrepresented in this study and no participants self-identified as Black. As a result, experiences may not fully represent the needs and experiences of people who use drugs in our setting.

CONCLUSION

Our findings demonstrate the various ways in which structurally vulnerable people who use drugs engage in risk reduction strategies within an increasingly toxic drug supply. While individual-level practices are readily utilized by participants, there is an urgent need to implement and scale-up community-level interventions to better support PWUD within the context of the current overdose crisis.