Self-reported pathways through which illicitly manufactured fentanyl enters the stimulant supply: Novel insights from people incarcerated for drug manufacturing and distribution

Abstract

Background:

The unintentional consumption of stimulants containing fentanyl among people who intend to only use stimulants contributes to overdose mortality in North America. Research exploring how fentanyl appears in the stimulant supply among people who manufacture and/or distribute drugs (PWDD) is critical to understanding supply-side factors that shape stimulant and opioid-involved overdose risk.

Methods:

From April to July 2023, thirty PWDD incarcerated at the Rhode Island Department of Corrections completed an in-depth interview about stimulant and opioid-involved overdose. Data were thematically analyzed to explore speculated pathways through which fentanyl may appear in stimulants.

Results:

Participants primarily endorsed unintentional fentanyl contamination pathways, including the accidental mix-up of drugs that look alike and cross-contamination via surfaces containing residual fentanyl where stimulant product is packaged. Congruent with historically contested beliefs about drug cutting and adulteration to induce dependence, some participants speculated that fentanyl may be intentionally added to stimulants to induce fentanyl dependence among people intending to only use stimulants to increase profits, though no participant reported firsthand knowledge of this. Participants extensively familiar with opioid and stimulant drug markets believed intentional contamination at high-level drug trafficking organizations was unlikely to occur as this would mix the drug markets, harming the profitability of maintaining distinct opioid and stimulant markets.

Conclusion:

Our findings challenge beliefs about the intentional addition of fentanyl in stimulants, showing them to be unsubstantiated. Instead, participants with advanced knowledge of drug manufacturing and distribution reported that fentanyl primarily appears in the stimulant supply unintentionally underscoring the need for targeted strategies to reduce contamination.

1. Introduction

The fourth wave of the United States (US) overdose crisis is characterized by a sharp increase in overdose fatalities that involve multiple substances (Friedman & Shover, 2023; Jones et al., 2020; National Institute on Drug Abuse, 2024). Of particular concern is a continual rise in overdose fatalities that co-involve stimulants, predominantly cocaine and methamphetamine, and illicitly manufactured fentanyl (referred to as fentanyl throughout) (Cano et al., 2020; CDC, 2024; Hoopsick & Andrew Yockey, 2023; Spencer et al., 2024, p. 491). Fentanyl is a highly potent synthetic opioid (Pichini et al., 2018), and its presence in the stimulant supply is driving an increase in stimulant and opioid-involved overdose deaths nationally (Friedman & Shover, 2023). Indeed, recent surveillance data approximate that 22,000 (~43 %) overdose deaths in 2022 involved opioids and stimulants together, outpacing mortality attributable to opioid overdose without the presence of stimulants (CDC, 2024). Although a rise in stimulant and opioid drug mixtures in the drug supply and associated overdose deaths has occurred nationally (Friedman & Shover, 2023; Park et al., 2021), stimulant and opioid-involved overdoses accounted for 44.3 % of all overdose deaths occurring in Rhode Island in 2022 (CDC, 2024). Though these data do not distinguish whether fentanyl was intentionally¹ or unintentionally² consumed with stimulants at the time of death, these data demonstrate that Rhode Island is heavily impacted by harms attributable to both fentanyl and stimulants.

The co-use of stimulants and opioids, whether intentional or not, is a documented risk factor for overdose (Liu & Singer, 2023), and may partially account for the rise in stimulant and opioid-involved overdose deaths in the US (Ciccarone, 2021; Kariisa et al., 2019). Research has begun to explore people who use stimulants’ (PWUS) use patterns to understand what role the intentional co-use of stimulants and opioids may have in shaping the fourth wave of the overdose crisis (Boileau-Falardeau et al., 2022; Hasgul et al., 2025; Mars et al., 2024; Ondocsin, Holm, et al., 2023). Motivations for concomitant stimulant and opioid use vary individually with some reasons including to enjoy the synergistic effect of both drugs types (Boileau-Falardeau et al., 2022; Ondocsin, Holm, et al., 2023), sequential use of both drugs types to manage effects of use (Boileau-Falardeau et al., 2022; Ondocsin, Holm, et al., 2023), and self-medication (Ondocsin, Holm, et al., 2023). Others have also articulated how large-scale changes to the drug supply, which impact drug availability and modes of use, have shaped the motivations to co-use stimulants and opioids (Mars et al., 2024).

Though the co-use of stimulants and opioids has increased in recent years, this rise does not explain opioid-involved overdose death among those who only intended to use stimulants prior to death. The unexpected presence of fentanyl in stimulant drugs heightens the risk of fatal overdose among people who only use stimulants as they lack an opioid tolerance (LaRue et al., 2019), are often less connected to harm reduction organizations, and may be unfamiliar with Good Samaritan Laws, making them less prepared to respond to an opioid overdose by using naloxone or calling 911 (Hughto et al., 2022). Our formative work with people who use cocaine in Massachusetts suggest that the inadvertent consumption of fentanyl in adulterated³ stimulants by people who intended to only use cocaine may contribute to the increase in stimulant and opioid-involved overdose deaths (Green, 2020). Disentangling the unanticipated presence of fentanyl in stimulant product from the rise in the intentional co-use of stimulants and opioids is needed.

It has been theorized by US agencies, including the Drug Enforcement Agency (DEA), that the rise in stimulant and opioid-involved overdoses may be a multifactorial consequence of drug trafficking organizations’ packaging, sale, and distribution practices that involve suppliers⁴ and people who manufacture and/or distribute drugs (PWDD)⁵ (Drug Enforcement Agency, 2021). Fentanyl may unintentionally contaminate⁶ the stimulant supply by way of cross-contaminated surfaces upon which multiple drugs are manufactured and packaged in a shared space (Drug Enforcement Agency, 2021), but how and the extent to which this may occur is unclear. Drug checking literature of stimulants has found a varied prevalence of fentanyl in US stimulant supplies, which gives some credence to this contamination theory. Among 718 stimulant samples obtained via a mail-in drug checking service in the US between 2021 and 2023, fentanyl was detected in 8.9 % of methamphetamine samples and 21.5 % of cocaine samples from 13 (of 25) participating states (Wagner et al., 2023). In 2023, 10 % (n = 94) of cocaine samples and 8 % (n = 44) of methamphetamine community drug checking samples obtained primarily in New England contained fentanyl (Streetcheck.org, 2024). In other instances, fentanyl has been detected seldomly in stimulants or not at all. In Philadelphia, just one methamphetamine, six cocaine, and four crack/cocaine samples submitted for drug checking in the first half of 2023 contained fentanyl (The Center for Forensic Science Research & Education, 2023). In New York City, an analysis found no fentanyl in 131 cocaine samples and 29 methamphetamine samples collected between 2021 and 2023 (Estrada et al., 2025). These data show that while fentanyl has been detected in stimulants, most stimulant samples remain fentanyl-free. The low detection of fentanyl in community-obtained stimulant samples raises questions about how fentanyl is sporadically appearing in stimulant products and may reflect contamination incidents that occur before drugs reach people for use. How supply side actions to create the stimulant drug supply result in the inadvertent consumption of fentanyl among people only intending to use stimulants remains unexplored.

Unsubstantiated claims suggesting that fentanyl is intentionally selected by drug manufactures as an adulterant in stimulants are prevalent during the fourth wave of the overdose crisis despite the lack of empirical evidence delineating the pathways through which fentanyl may appear in stimulant drugs. Unsubstantiated assertions about fentanyl’s presence in stimulants reflect commonly expressed beliefs regarding manufacturing and distribution practices that are predicated on the assumption that all PWDD are nefarious. For example, it has been speculated that fentanyl is added to stimulant product to induce fentanyl dependence among PWUS thereby increasing a drug distributor’s profits by expanding the number of people dependent on fentanyl; this is a contemporary reformulation of the historically pervasive pusher myth whereby PWDD deceptively sell (i.e., push) an adulterated drug product to hook (i.e., establish substance dependence) unsuspecting customers to adulterants to boost drug sales (Coomber, 2006). The assumption of intentional adulteration of stimulant with fentanyl has a historical basis played out in older narratives, such as the claim that in the 1990s and 2000s ecstasy was commonly adulterated with heroin (Coomber, 2006) or the claim that adulteration of the US heroin supply of the 1990s was commonplace, despite having not been substantiated by forensic analysis (Coomber, 1997, 1999). Contemporary media coverage of opioid overdose events among those who use stimulants often resurface assumptions of systematic adulteration practices. For example, the accusation that fentanyl-adulterated stimulants are intentionally sold to PWUS has been reported by US Attorneys when discussing these deaths despite no conclusive evidence to suggest an intentional act on the part of the PWDD to intentionally sell adulterated stimulants (Harper, 2023). The media has also reported that PWDD have resorted to using fentanyl to bulk up their inventory of drugs to have a continual supply of drugs to sell (Nir Maslin, 2021) – an unsubstantiated assertion of dilution practices that is incompatible with international efforts to curb the availability of fentanyl and implies, without forensic evidence, that fentanyl is a common cutting agent in stimulants (The Global Coalition to Address Synthetic Drug Threats, 2024). In fact, a review of forensic evidence of stimulant product determined that adulteration typically utilizes various substances that are inert, mimic or enhance the effect of an unregulated substance, or facilitate administration of the substance, contrary to presumptions about intentionally harmful adulteration practices (Cole et al., 2010). Historical and ongoing forensic analysis of unregulated stimulants has not provided evidence that intentional cutting of stimulants with fentanyl to either induce dependence or dilute for profit is an evidenced practice of PWWD in North America.

Despite varied theorizations about how fentanyl may appear in the stimulant supply and the contemporary emergence of historical drug adulteration claims, there has been no concerted research effort to explore how and why fentanyl sporadically appears in the stimulant supply from the perspective of PWDD. People involved in the manufacturing and distribution of drugs have insights into the supply, manufacturing, distribution, and trafficking of the fentanyl and stimulant drug markets. As such, the perspectives of PWDD are needed to interrogate the issue of fentanyl’s sporadic manifestation in forensic analysis of stimulants and mortality data as well as to provide clarity amidst the backdrop of speculation about how fentanyl appears in the stimulant supply. Thus, this qualitative study is in direct response to the need for supply-side focused research to understand pathways through which fentanyl may appear in the stimulant supply as characterized by participants with extensive expertise in manufacturing and/or distributing drugs who were incarcerated at the Rhode Island Department of Corrections (RIDOC).

2. Methods

2.1. Overview and design

As part of a parent study to identify risk and protective factor for stimulant and opioid-involved overdose(Hughto et al., 2024), between April and July 2023 30 people who were incarcerated at the RIDOC for drug manufacturing and/or distribution completed an approximately 1 h study visit consisting of two parts: an approximately 20-min interviewer-administered electronic survey immediately followed by an in-depth audio recorded interview that ranged in length from 16 to 52 min to explore how and why fentanyl enters the stimulant supply among other topics. Eligible participants were [1] ≥ 18 years of age; [2] able to speak and understand English or Spanish; [3] sentenced for manufacturing and/or distributing illicit drugs; [4] incarcerated for ≤ three years; and [5] willing and able to provide informed consent. Participants awaiting trial were excluded given the sensitive nature of the research question. This research was reviewed and approved by the Brown University Institutional Review Board and the RIDOC Medical Research Advisory Group.

2.2. Recruitment

RIDOC administrators provided a list of incarcerated individuals who met eligibility criteria 1, 3, and 4 to the senior author. The list included the individuals’ current and previous charges. The senior author selected potential participants who were diverse in terms of age, race/ethnicity, gender, and who had charges related to cocaine manufacturing and delivery. Potential participants were mailed, to their RIDOC mailbox, a letter that described the study and contained study contact information. Approximately seven business days after mailing the letter, study staff traveled to the RIDOC to meet with potential participants. Correctional officers informed potential participants of the team’s arrival and asked them if they would like to meet with a member of the team in a private visiting room to learn more about the study. Sixty-one potential participants received a letter, of which 13 (21 %) were released or could not be contacted. None of the potential participants who were invited to meet with us, and were available to meet, declined to do so. After learning about the study, 14 (23 %) declined to proceed to eligibility screening, and four (7 %) could not proceed due to challenges in aligning potential participants’ schedules with the availability of our Spanish speaking researcher. Thirty participants (49 %) were eligible, consented, and completed the study visit.

2.3. Data collection

Participants provided written informed consent prior to participation; the consent process included informing participants about protections afforded by a Certificate of Confidentiality. Study visits were conducted by one of three trained members of the study team with expertise in substance use research and qualitative methods. Participation consisted of an interviewer-administered electronic survey immediately followed by an in-depth audio-recorded interview that explored survey domains in greater depth. Interviewers encouraged participants to speak generally (i.e., not from the first-person perspective) during interviews about drug manufacturing and distribution practices to minimize the risk of self-incrimination for activities unrelated to their current sentence, although ultimately, participants described their personal knowledge for most questions. Interviewers completed a post-study visit memo to document observational and theoretical notes regarding the survey and interview and to facilitate our conceptualization of emergent themes (Birks et al., 2008). Participants received $40.00 in their commissary account for participation.

2.4. Survey measures

Age (in years), race/ethnicity (combined to white, non-Hispanic; Black, non-Hispanic; Native American, non-Hispanic; Hispanic; and multi-racial, non-Hispanic), and sex assigned at birth (male, female) were assessed. Educational attainment was coded as high school or less; high school or General Education Development (GED) equivalent; and some college or more. Estimated weekly income from all sources prior to incarceration was assessed in US dollars. Lifetime and past-30-day substance use including powdered or crack cocaine; methamphetamine or amphetamines; fentanyl; heroin; or real or fake prescription opioids was also assessed.

Participants were also asked about the length of their current sentence and the date they were incarcerated, which was used to stratify participants according to whether they had been incarcerated for more or less than 6 months at the time of enrollment. Lifetime history of arrest for possession; selling or intent to distribute; trafficking; and manufacturing illegal drugs or paraphernalia was also assessed. We assessed drugs involved in participants’ last conviction rather than broadly asking about drugs that were distributed prior to incarceration or associated with prior charges as we did not want to incriminate them for charges that did not result in conviction. Substances involved at last arrest included heroin; fentanyl; real and fake opioid pain medication; crack cocaine; powdered cocaine; methamphetamine; real or fake amphetamines; methadone; buprenorphine; non-opioid pain medications; benzodiazepines; ecstasy; GHB; hallucinogens; ketamine; and cannabis. To explore whether beliefs regarding how fentanyl may appear in the stimulant supply differs by participants’ drug market familiarity, we categorized drugs involved the participants’ convictions as: opioids only (i.e., fentanyl or heroin); stimulants only (powdered or crack cocaine, methamphetamine); both opioids and stimulants; or cannabis. Cannabis was a specified category because three participants’ last conviction reportedly only involved cannabis distribution though they had evidence of past cocaine-related charges in their administrative data. In all cases, individuals whose last conviction did not involve cocaine voluntarily disclosed extensive knowledge of stimulant and opioid markets based on their experiences with manufacturing and/or distribution unrelated to their current sentence for cannabis and were therefore retained in the study sample.

2.4.1. Awareness of fentanyl in the drug supply

Four variables assessed participants’ familiarity with the presence of fentanyl in the drug supply. Item one asked participants to indicate if they had heard of people selling drugs that contained fentanyl; participants who responded “yes” were asked to specify which drugs they were being sold as. If powdered cocaine, crack cocaine, or methamphetamine were reported, they were considered to have heard of people selling stimulants that contained fentanyl. Item two asked if people who sell drugs ever unintentionally put fentanyl into drugs and item three asked if people who sell drugs ever intentionally put fentanyl into drugs. Lastly, item four asked whether people who sell drugs ever unintentionally sell or distribute the wrong drugs (e.g., fentanyl is accidently sold as cocaine). The response set for the four items included yes; no; I don’t know; and prefer not to answer.

2.4.2. In-depth interview topics

Interviews were guided by a topic guide that explored survey items to greater depth. Topics relevant to this analysis explored perceptions of how fentanyl appears in drug supply, and we probed about stimulant drugs specifically as well as general knowledge about relationships between drug suppliers, PWDD, and people who use drugs (PWUD).

2.5. Data analysis

Quantitative analyses were conducted in SPSS version 29.0.10. Items were assessed for missingness; two participants did not provide a length for current sentence so the analytic sample for this item was 28. Descriptive statistics (means, medians, and frequencies) were calculated for all quantitative variables. Interviews were professionally transcribed, and transcripts were assigned a randomly generated pseudonym and reviewed for accuracy by the study team. Transcripts were coded by three trained coders (who also collected the data) using Dedoose Version 9.0.17 and analyzed following an integrated approach to thematic analysis in five phases (Bradley et al., 2007). An audit trail maintained by the first author was used to document coding and analysis decisions. In the first phase, a preliminary codebook was created by identifying deductive codes from two sources: (1) a codebook used in the parent study to code interviews with PWUS, and (2) the topic guide with PWDD. In the second phase, the first two transcripts were selected for an independent line-by-line open coding process by the coders. During this process, the draft codebook was iteratively refined to add inductive codes for emergent concepts not represented in the existing codebook, co-locate conceptually related codes and revise code names and definitions, and collapse codes to reduce redundancy. Following these revisions and for the third phase, six (20 %) transcripts were randomly selected to be independently double coded by two coders. Reliability in the frequency of code application across these six transcripts reached 81 %; discrepancies in code application were resolved through discussion, and a third coder was consulted to achieve consensus as needed. The remaining 24 transcripts were independently coded, with each coder coding 8 transcripts. The coding team met weekly through the end of the coding process to discuss emergent themes.

The fifth and final stage of analysis was led by the first author. A subset of codes relevant to answering the research question was activated in Dedoose and excerpts were exported into a separate document; each excerpt was tagged with the drug category (e.g., stimulants only) involved in the participants’ last arrest, age, and sex. This exploratory analysis focused on understanding PWDDs’ theorized pathways through which fentanyl may appear in the stimulant supply, with attention to how substances dealt at last arrest and participants’ drug market familiarities (i.e., direct firsthand knowledge vs. hypotheticals and speculation) color these perceptions. After characterizing the pathways, the Haddon Matrix (Haddon, 1970, 1980), a heuristic framework often utilized in injury prevention research to consider the multi-level risk and protective factors before, during, and after an injury or death, such as fatal overdose (Barnett et al., 2005; Runyan, 1998), was used to aid the interpretation of the results. Specifically, we leveraged the Haddon Matrix to understand how the thematically identified, theorized pathways of fentanyl exposure may introduce the risk of stimulant and opioid-involved overdose. Pathways are discussed in the context of general assumptions that have been made about fentanyl’s presence in stimulants. Situating the findings in both a risk framework and historical context helped to inform the intervention and policy implications that are presented in the discussion.

3. Results

Participants were on average 35 years old (range: 21–53) and mostly men (n = 26, 86.7 %). Nearly half were Hispanic ethnicity (n = 13, 43.3 %), one third were non-Hispanic people of color (n = 9, 30 %), and most had at least a high school degree or GED (n = 22, 73.3 %). See Table 1 for additional demographic characteristics. The majority (n = 25, 83.3 %) of participants had been incarcerated for seven months to three years, and 12 (40 %) had distributed drugs within the year prior to participation. Participants reported an estimated median pre-incarceration weekly income of $3250 (IQR $1612.50 - $6,250, roughly $83,850 - $325,000 yearly), suggesting that our sample did not include low-income earners or multi-million-dollar earners within drug trafficking organizations. The median sentence length for current sentence was five years. Most participants had been arrested at least once in their lifetime for selling (n = 29, 96.7 %), possession (n = 28, 93.3 %), and delivery (n = 28, 93.3 %) of illegal drugs or drug paraphernalia. Twenty-four participants (80 %) had been arrested for manufacturing illegal drugs, whereas fewer (n = 8, 26.7 %) had been arrested for trafficking illegal drugs or drug paraphernalia. Drugs involved in participants’ last arrest varied and included stimulants and opioids (n = 14, 46.7 %), stimulants only (n = 12, 40.0 %), cannabis only (n = 3, 10 %), and opioids only (n = 1, 3.3 %).

Table 1.

Participant demographics (N = 30).

Participant characteristic	N(%)
Age
Mean	35 (range: 21–53)
Sex
Male	26 (86.7)
Female	4 (13.3)
Race and ethnicity
Hispanic	13 (43.3 %)
White, non-Hispanic	8 (26.7 %)
Black, non-Hispanic	5 (16.7 %)
Multiple races, non-Hispanic	3 (10 %)
Native American, non-Hispanic	1 (3.3 %)
Educational Attainment
Some high school or less	8 (26.7 %)
High school degree or GED	14 (46.7 %)
Some college or more	8 (26.7 %)

3.1. Speculated unintentional pathways through which fentanyl enters the stimulant supply

Most (n = 24, 80 %) participants reported that they had heard of people selling stimulants that contained fentanyl. Of 29 responses, 19 (65.5 %) had heard of cocaine, crack cocaine (n = 12, 41.4 %), amphetamines (n = 8, 26.7 %), and methamphetamine (n = 7, 23.3 %) being sold with fentanyl. Most participants felt that these drug combinations are most commonly happening because of the mix-up of fentanyl with powdered cocaine similar in appearance and the cross-contamination of workstations used by PWDD who manufacture and distribute stimulants and fentanyl.

3.1.1. Unintentional pathway A: accidental mix-up of fentanyl with stimulants

Many participants (n = 20, 66.7 %) believed that PWDD may accidently sell the wrong drug, such as when fentanyl is accidently sold as cocaine. Fentanyl and stimulant mix-ups are reportedly more likely to happen when PWDD “move too fast” when selecting a product to make a sale to a PWUS. ‘Luke’ shared:

Sometimes you run around and you baggin’ ’em all up and you don’t get a chance to do that [mark the bag with the client’s initials]. Then you go to meet somebody, you might give ’em the wrong bag. You might give fentanyl to the person that wanted coke and coke to the and then you have to call ’em real quick and fix that before they sniff a line of coke not realizin’ it’s fentanyl and you gonna die. [stimulant and opioids, 44, male]

Rushing when distributing drugs was characterized as a “sloppy” business practice because it increases the likelihood of confusing one drug for another; these mix-ups introduce the potential for legal and lethal consequences. The risk of these mix-ups is particularly heightened when the PWDD distributes both stimulants and opioids because the products can appear similar. For example, ‘Tristan’, who sold stimulants and opioids, described how he accidently sold powder cocaine to his roommate who regularly purchased fentanyl, resulting in a stimulant high:

I was happy that obviously she didn’t die. But then I gave her the fentanyl and she kinda calmed down, and then we laughed about it. And I was like, ‘oh, I apologize, I couldn’t tell the difference.’ The fentanyl I had has many different shades. Some of it was brown, most of it was white, some of it was blue, some of it was pink. At that time, the one I had was white. [stimulants and opioids, 30, male]

Although the participant’s roommate, who used fentanyl, unintentionally consumed a stimulant because of a bag mix-up, this exemplifies the potential harm of conflating powdered fentanyl with powdered cocaine. Among people who only use stimulants, who lack a physiological tolerance to opioids, and who are less prepared to navigate an opioid overdose, this type of accident can be fatal. ‘Jermaine’, a 25 year-old-male shared: “You could easily have a couple bags of stuff that looks the same, couple bags of white stuff the same, and end up, oh someone calls you out, ‘I need this,’ and you just take it out the wrong bag.”

Participants also expressed that street-level PWDD who intend to sell stimulants may occasionally unknowingly sell stimulants containing fentanyl because they have limited knowledge or control over the contents of the drugs purchased from a mid-level distributor to sell on the street. For example, ‘Elias’, a 26-year-old man whose arrest involved stimulants, theorized that stimulant and opioid-involved overdoses may sometimes be the result of a street-level PWDD inadvertently distributing a stimulant containing fentanyl: “The only time I think that it’s accidental is if you don’t know that it’s [fentanyl] in there, and you buy it and sell it to somebody else. That’s probably the thing I can think about it being accidental.” Here, a nuance to understanding how accidental sales of fentanyl to people only intending to use stimulants occur is evidenced and the importance of understanding who delivers the drug product (e.g., a runner vs. mid-level distributor) and their understanding of the contents of the drugs they sell is clear. Limited knowledge of the drug supply among street-level PWDD underscores the need for harm reduction strategies that inform street-level PWDD about the content of their supply to reduce instances of the unintentional distribution of stimulants containing fentanyl to people intending to only use stimulants.

3.1.2. Unintentional pathway B: cross-contamination via surfaces containing fentanyl Residue

Participants acknowledged the potential for fentanyl to unintentionally enter the stimulant supply because of accidental cross-contamination of a substance with residual substances on surfaces and tools used when manufacturing and preparing drugs: “You just weigh something and, uh, weigh something else, and that little bit gets into whatever else you weigh.” [‘Nichole’, stimulant and opioids, 33, female]. One participant, ‘Omer’, who distributed stimulants and opioids, likened not cleaning a workstation in between packaging stimulants and opioids to “how E. coli, or Salmonella from chicken or something” contaminates food. ‘Cederic’, whose last arrest involved stimulant distribution, attributed this to “laziness” and PWDD “trying to cut corners” by not cleaning off weighing stations or plates, indicating that the risk of cross-contamination is modifiable.

Participants articulated how drug use by PWDD may explain why cross-contamination of stimulant product with fentanyl and the unintentional mix-up of fentanyl with stimulant product occur. For example, ‘Ian’, attributed unintentional errors that brought fentanyl into the stimulant supply to distributing drugs while high:

Some [PWDD] are using it the same time as dealing. So, they’re mixing up their products, or they’re not labeling what substance is in what bag or they’re using the same instruments to weigh and measure. It’s cross-contaminating everything. It’s [fentanyl] mixing in with substances that it shouldn’t be. Fentanyl is being mixed with cocaine … it’s unintentional, like I said, from the fact that people, just, they’re not paying attention to what they’re doing. They get confused. [‘Ian’, stimulant and opioids, 31, male]

Here, ‘Ian’ acknowledges that PWDD may use their own supply; in fact, all but one participant had a lifetime history of any illicit drug use (97 %), and half had used any illicit stimulant or opioid in the 30 days prior to incarceration. Drug use can impair the clarity needed to maintain orderly manufacturing operations and accurate distribution operations. Relatedly, other participants, like ‘Cederic’ attributed accidents to exhaustion in addition to drug use: “being up for days not paying attention. Just carelessness. Smoking too much weed or whatever, any type of drug”. Collectively, exhaustion and using drugs while working constitutes a risk environment in which unintentional errors like distribution mix-ups and cross-contamination resulting in fentanyl’s presence in stimulants occurs.

3.2. Speculated motivations to intentionally add fentanyl to the stimulant supply

Nearly all participants (96.7 %) believed that PWDD intentionally put fentanyl into drugs. For example, twenty-four (80 %) participants reported that they had heard of pain medications (e.g., oxycodone) containing fentanyl, which aligns with forensic evidence indicating fentanyl has supplanted heroin and other unregulated opioids in the US. Participants shared that fentanyl is used because fentanyl is cheaper, more accessible, and produces an even greater “downing” effect than other opioids like heroin or prescribed opioids. While no participant reported intentionally adding fentanyl to stimulant products, most (n = 24, 80 %) participants had heard of this happening. Yet when asked why fentanyl might be intentionally added to the illicit stimulant supply, participants shared unsubstantiated claims about stimulant adulteration that are common in the media and reflective of historical beliefs about drug adulteration practices rather than first-hand accounts of intentional adulteration of stimulant product with fentanyl.

3.2.1. Speculated intentional pathway A: adding fentanyl to stimulants to induce dependence

Some participants, like ‘Nichole’ whose last arrest involved stimulants and opioids, unwittingly perpetuated the narrative of the ‘pusher’ distributor (Coomber, 2006) wherein PWDD intentionally add fentanyl to stimulants to establish a dependent customer base. ‘Nichole’ shared “So that you [PWUS] catch a habit. Physical addiction.” It was stated that because stimulants do not cause physical dependency, but rather a “mental one,” adding fentanyl to stimulants could be a way to get PWUS physically dependent on fentanyl in stimulants, ultimately generating greater profits. ‘Daryl’ crafted a hypothetical situation to convey this idea:

They [PWDD] could just be like ‘I [PWDD] wanna hook that person that does the coke onto fentanyl so I’m [PWDD] just gonna give ‘em [PWUS] a little bit of fentanyl so that way they [PWUS] ask for the fentanyl instead when they [PWUS] still feeling sick with the cocaine […] I’ma [PWDD] hook ‘em using this drug slowly. [stimulant and opioids, 28, male]

The belief that PWDD intentionally add fentanyl to the stimulant supply to induce physical dependence was presented without firsthand experience of participants who had done this or who knew of people doing this. This unsubstantiated belief was also more commonly endorsed by participants who had distributed stimulants but not opioids prior to arrest, like ‘Anita’, a 34-year-old woman. She shared:

Most of the bigger dealers purposely put it in [fentanyl into stimulants]. It’s a bigger money maker for them. More people come back because now they’re addicted to it. Coke and crack is not really an addictive drug, it’s more of a mental thing. However, if they put an addictive drug in it …

Other participants offered perspectives informed by their experiences distributing and using stimulants (n = 22, 73.3 %) that contrasted the belief that PWDD intentionally add fentanyl to stimulants and identified this belief as an illogical misconception. Selling a fentanyl-adulterated stimulant is counter to what people who only use stimulants expect when purchasing stimulants. For example, ‘Johnnie’, a 33-year-old male who only dealt stimulants prior to arrest but also reported using stimulants said: “I don’t want to sniff coke and then just fall off with fentanyl. Like, what the fuck? Just smokin’ blind, just die, like, that’s crazy.” Likewise, ‘Tristan’ had a keen understanding of drug use preferences because of his own familiarity with using stimulants and opioids separately for different purposes: “If I wanted to feel good, I would sniff the pills or whatever. But if I needed to stay up because I was on drugs, then I would sniff cocaine. So, I would have both in my system. I didn’t do it at the same time.” [stimulant and opioids, 30, male]. ‘Johnnie’ and ‘Tristian’s’ perspectives are informed by their direct experiences using the drugs that they also sell; it is this firsthand knowledge that discredits the distributor ‘pusher’ narrative shared by participants like ‘Daryl’, whose views on intentional adulteration of stimulants with fentanyl were speculative and not supported by direct or witnessed experience.

Participants, like ‘Eddy’, a 27-year-old participant whose last arrest only involved stimulants, believed that intentionally distributing stimulants that contain fentanyl may upset clientele who intend to use fentanyl-free stimulant product:

I’ve never met anyone - and not to have a big head, like, but I’m not a street dealer - you wouldn’t - I’ve never heard anyone, “Hey, man, like, what do you think about this?” Or like, “Oh, this guy’s been cutting his coke with fentanyl and -” Just don’t do it ‘cause the person who wants something, they want what they want.

Here, ‘Eddy’ attempts to establish the credibility of his beliefs by distancing himself from “street dealers” to convey his advanced knowledge of the drug supply and to emphasize that people who manufacture drugs would not cross the stimulant and opioid markets because that is not desired by people who use stimulants.

3.2.2. Speculated intentional pathway B: adding fentanyl to stimulants to increase profitability

Some participants speculated that fentanyl may intentionally be added to stimulants by street-level distributors to increase their ability to make money, which is a regurgitation of the ‘pusher’ narrative (Coomber, 2006). When asked at what level in the drug supply fentanyl is added into the stimulant product, ‘Ian’ offered that street-level distributors may be motivated by profits to add fentanyl into stimulants they are selling:

It’s the lower-level dealers that are adding it [fentanyl] in … If they’re going to cut it, they’ll do it themselves. So, that’s always a lower-level issue … Whatever they can do to, stretch it a little bit. That’s a lot of where that is. It’s the lower guys with it. It’s the little guy, the very bottom. [stimulants and opioids, 31, male]

‘Pasquale’, a 49-year-old man who was involved in distributing stimulants and opioids prior to arrest and who had extensive experience with drug markets, shared ‘Ian’s’ speculation that the addition of fentanyl into stimulants, if happening intentionally at all, would most likely happen at the street level by distributors motivated by profit:

I think it’s more on the lower end who’s just trying to make extra money that’s just trying to stretch it out. I would probably say there’s probably like lower-level dealers to runners who’s actually doing it.

These participants offer no direct experience with the intentional adulteration of stimulants with fentanyl and instead construct a view of what they speculate could happen at the street level by the hypothetical “guy on the street …” who thinks that the intentional addition of fentanyl to stimulants is lucrative. Here, a narrative that intentional adulteration of stimulants with fentanyl occurs at the street level is offered, rendering the claim that street-level distributors adulterate stimulants with fentanyl unsubstantiated.

At the same time, when reflecting on whether fentanyl was added to stimulants at the high-level drug trafficking organizations, these participants speculated that this is unlikely to occur and grounded this in the knowledge that people who distribute both opioids and stimulants can maximize profit by maintaining separate markets for stimulants and opioids. ‘Pasquale’ explained:

I don’t really see them [cartels] doing it that way [adulterating stimulants with fentanyl] because that’ll just mess up the nucleus of what business is. You’ll just cross product. You’ll lose your distribution. You’re not going to make the same amount of money because you want one-to-one and one-to-one; keep it separate. One’s fentanyl, one’s heroin, one’s cocaine, you know.

Here, ‘Pasquale’ conveys his working knowledge of the drug market, and a sentiment shared by participants who had equivalent levels of experience with stimulant and opioid drug markets: systematic adulteration of stimulants with fentanyl is improbable because it would be bad for business. As ‘Ian’ explained: When you deal with wholesalers [suppliers], they’re not cutting their drugs with fentanyl. When you’re dealing with weight, people want pure product. They don’t want nothing cut. [stimulants and opioids, 31, male]. Our participants with years-long experience manufacturing and distributing stimulants and opioids at a high-level characterized suppliers as striving to move quality product because they have placed great financial investment into operations. In other words, it is a nonsensical financial decision for suppliers of large quantities of product to sell fentanyl adulterated stimulant product.

4. Discussion

To our knowledge, this is the first study of incarcerated people with advanced knowledge of drug manufacturing and/or distribution practices to characterize how fentanyl may appear in the stimulant supply. Through recollections of personal and witnessed experiences, participants largely attributed the presence of fentanyl in the stimulant supply to human error, including the accidental mix-up of fentanyl with stimulants while packaging and distributing drugs and the cross-contamination of surfaces containing residual fentanyl while manufacturing, all of which are exacerbated by possible co-use of substances during operations. Conversely, no participant reported engaging in intentional adulteration of stimulants with fentanyl, and participants’ perspectives regarding how this may happen were largely speculative and reflective of historical drug adulteration myths. For example, some participants speculated that street-level distributors add fentanyl to stimulants to increase their opportunity to make money despite having no firsthand knowledge of this occurring, rendering this claim unsubtantiated. But participants who used drugs and who had extensive expertise distributing and/or manufacturing drugs refuted this as improbable because it is more lucrative to maintain separate stimulant and opioid markets. Collectively, these findings provide empirical evidence demonstrating that the most probable way fentanyl appears in the stimulant supply is through human error, rather than the intentional adulteration of stimulants with fentanyl, suggesting that interventions to reduce accidental contamination of the stimulant supply with fentanyl are needed.

Our findings challenge persistent assumptions about the intentional addition of fentanyl in stimulants, showing them to be unsubstantiated. Some participants whose arrest involved distribution and/or manufacturing of stimulants speculated that PWDD may use fentanyl as an adulterant to stimulants to induce opioid dependence among PWUS, thereby creating a profitable customer base. This is a contemporary manifestation of a long-standing belief that people who manufacture drugs intentionally add opioids to non-opioid drugs to induce dependence, when in reality the time from first use to dependence is not instantaneous and instead resultant from sustained use (Coomber & Sutton, 2006). As just one historical example of this dependency myth, in 1993, reports circulated that heroin was intentionally being added to ecstasy in London despite no drug surveillance data to support this assertion (Coomber, 2006). This is not unlike modern DEA and media reports of stimulants that are purportedly “laced” with fentanyl (Daly, 2021; Ordonez & Salzman, 2023). Some participants in this research speculated that the phenomenon of inducing dependence may be at play with respect to whether fentanyl is intentionally added to stimulants, even though no participant recollected personal or witnessed experience with the intentional adulteration of stimulants with fentanyl. Coomber’s (1997) seminal qualitative research with London-based PWDD also found that participants believed potentially harmful adulterants were being used in the drug supply despite most participants having no firsthand experience doing so. Like Coomber (2006), some participants in this research reiterated beliefs about how fentanyl appears in stimulants that are consistent with societal misunderstandings of this phenomenon, despite that their own experiences do not support the assumption that fentanyl is intentionally added to stimulants. At the same time, participants whose last arrest involved both stimulants and opioids and who had extensive firsthand manufacturing and/or distributing experience discredited the belief that fentanyl is intentionally used to adulterate the stimulant supply because this is antithetical to the business of maintaining separate drug markets and that there is a high risk of fatal overdose to unsuspecting and potentially fentanyl naïve PWUS (Liu & Singer, 2023). Therefore, the idea that fentanyl is intentionally used to adulterate stimulants is not only unsupported by our data but is also considered illogical by participants with direct firsthand knowledge of drug manufacturing. Future research conducted in non-carceral settings in diverse localities should recruit PWDD across levels of drug market hierarchies to explore how variability in knowledge about adulteration and contamination is shaped by participants’ drug market roles. Additionally, the notion of maintaining separate drug markets should be further interrogated amidst the rise in polysubstance use across the US.

Our study found that the most probable pathway through which fentanyl appears in the stimulant supply is because of human error. Indeed, participants in this study articulated that fentanyl accidently appears in the stimulant supply by multiple pathways including cross-contaminated surfaces in clandestine environments where multiple drugs are manufactured and packaged in a shared space, during preparation for sale by people who manufacture both opioids and stimulants while they themselves are under the influence of these substances, and through bag mix-ups more likely to occur in an unregulated environment where quality control and consistent labeling are absent. Our findings support the beliefs of harm reduction workers that suggest fentanyl’s presence in the stimulant supply is accidental, occurring during the manufacturing and distribution process (Ondocsin, Ciccarone, et al., 2023). This unintentional pathway also helps to contextualize the variable prevalence of fentanyl in stimulant samples in Canada and the US. For example, of 110 expected stimulant samples collected in Toronto from 2019 to 2020, 4 % (n = 4) contained an opioid including fentanyl (Scarfone et al., 2022). In New England, the drug market our participants played an integral role in, drug checking services have detected a slightly greater prevalence of fentanyl in cocaine (0 %–23 %) and methamphetamine (0 %–38 %) between 2020 and 2024 (Streetcheck.org, 2024). And while the DEA has seized stimulants containing fentanyl (Drug Enforcement Agency, 2018a, 2018b), the quantity of these stimulant-fentanyl mixtures is small relative to seized stimulants that do not contain fentanyl. If there was a large-scale operation to adulterate stimulants with fentanyl underway, the percentage of seized stimulant-fentanyl combinations and the percentage of community-obtained stimulant samples testing positive for fentanyl would likely be higher. Variability in the detected fentanyl prevalence in the North American stimulant supply reflects differences in drug markets and supports our participants’ theorization that fentanyl most often unintentionally appears in stimulants rather than being a result of intentional adulteration. Future research should more fully characterize the manufacturing practices of people who experiment with mixes and drug cuts, particularly among those working in stimulant and opioid drug markets, to further explore how contamination occurs, which stimulants these accidents are more likely to occur with (e.g., powder cocaine vs. crystal meth), and to identify techniques to reduce these errors.

Participants in our study believed that the accidental sale of fentanyl to people intending to only use stimulants is contributing to stimulant and opioid-involved overdoses. These mix-ups have resulted in the death of PWUS after the inadvertent use of fentanyl, as reported in the media (Ordonez & Salzman, 2023; Patrick, 2022; Petrishen, 2022). We heard from participants that these type of distribution accidents are more likely to happen if the distributor is intoxicated. Most of our sample also used drugs prior to incarceration, bolstering the validity of this perception and reiterating that there is overlap between distributors and PWUD. Relatedly, some participants speculated street-level distributors’ lack of knowledge about the drug product they sell may play a role in the accidental distribution of incorrect product, but who these individuals are (e.g., runners vs. mid-level distributors) and to what extent these individuals account for these accidental sales is not clear from our data and should be explored further. Errors like the unintentional sale of the wrong product may undermine the trust that PWUD place in PWDD to provide the correct product (Bardwell et al., 2019; Carroll et al., 2017, 2020). A breach of trust may hinder open conversations that foster a mutual understanding of the contents and strength of the drug supply, which is important to help people make informed decisions about how they use their drugs.

Our finding that participants believe fentanyl predominantly enters the stimulant supply unintentionally by human error suggests that strategies are needed to reduce errors and improve manufacturing processes. Accordingly, it is worthwhile to explore the direct involvement of PWDD in interventions to reduce harms to PWUD given their large role in shaping drug markets and use practices. Other work with PWDD has reported interest in harm reduction strategies like drug checking and fentanyl test strip adoption to improve the consistency of drug product and reduce harms to PWUD (Betsos et al., 2021; Hedden-Clayton et al., 2024). It may be worth exploring expanded utilization of standardized packaging to differentiate between opioid and stimulant products (e.g., clearly distinguish powder cocaine from fentanyl) to reduce incidents of confusing these drugs for each other. Whether PWDD would be receptive to this practice remains to be understood because branding product could potentially be used by law enforcement to trace a product back to its distributor, raising concerns of criminal prosecution. Developing peer-delivered educational workshops for PWDD about manufacturing practices that would reduce harms to PWUD is warranted and would also serve to mitigate the risk that PWDD are charged with drug-induced homicide crimes. To that end, a US based qualitative evaluation of six PWDD who participated in a harm reduction training tailored to PWDD indicated that this population had high interest in tips for how to discuss the contents of their drug supply with customers and the use of fentanyl test strips with the people that they sell to (Hedden-Clayton et al., 2024). A future direction of this work could be to replicate such training and incorporate information about safer-manufacturing practices to reduce human errors. This supply-side focused training would be novel, have the potential to be high-impact given PWDD’s role in shaping the drug supply, and be a needed complement to safer-use trainings for PWUD.

The use of drug-checking services by PWDD should also be explored. To date, the majority of drug-checking services have been utilized by PWUD to test their drugs (Giulini et al., 2023), but these services could have an even wider protective effect if adopted by PWDD given the role they play shaping the drug supply (Bardwell et al., 2019). Drug-checking services could be utilized to inform distributors about the contents of their product, helping to reduce sales of contaminated product before the point of sale, ultimately reducing harms to PWUD. For example, The Vancouver based Drug User Liberation Front’s Compassion Club and Fulfillment Centre once operated a novel tested drug distribution network (Nyx & Kalicum, 2024), but ceased operations due to criminalization. Indeed, the looming threat of criminalization complicates systemic integration of quality assurance practices; trafficking, distributing and/or manufacturing fentanyl is highly criminalized, and people who have sold substances that contain fentanyl involved in overdose deaths have been convicted of manslaughter because of drug-induced homicide laws (Beletsky et al., 2022; Morrissey et al., 2024; Park et al., 2020). Criminalization has meant that drug production operations largely forgo quality assurance practices that could reduce instances of cross-contamination and subsequent unintended harms to PWUS like stimulant and opioid-involved overdose (Cole et al., 2011). This legal reality makes it challenging to implement needed supply-side focused harm reduction efforts to respond to the overdose crisis, such as drug checking, and underscores a need to adopt a health policy and harm reduction lens when revising laws affecting the drug market.

There are several limitations to this work. First, the experiences and perceptions of participants reflect the drug market in Rhode Island and are not directly transferable to other regions of the world where drug markets differ. Second, we did not directly assess organized crime and instead relied on participants’ characterizations of how their experience distributing and/or manufacturing drugs compared to other participants to understand their relative positioning in the drug market hierarchy. Consequently, there is a lack of information regarding drug trafficking organizations beyond the involvement of suppliers, PWDD and PWUD. Third, because the survey did not assess lifetime history of drugs involved in distribution and/or manufacturing, we were limited in our ability to fully understand participants’ drug market familiarity beyond the drugs involved in their last arrest. Relatedly, for ethical reasons we could not assess potential participants’ experiences with drug manufacturing and distribution prior to determining eligibility and obtaining consent. So, we treated convictions from participants’ last arrest as a proxy for practical knowledge of the drug markets to determine study eligibility. Convictions may not necessarily equate to firsthand experience in all cases. Other investigators using conviction history to determine study eligibility should systematically measure knowledge and experience with drug manufacturing and distribution to validate this approach. Fourth, given that we spoke with people who were incarcerated, and therefore not actively engaged in drug manufacturing and/or distributing, it is possible that the present characterizations differ from the current state of the drug market. None-theless, this evidence specific to stimulant drug markets is needed and demonstrates that it is possible to engage incarcerated people in research about the drug supply. Fifth, it may also be true that participants avoided sharing beliefs they perceived would be unfavorable or put them at risk of further criminalization, such as the intentional adulteration of stimulants with fentanyl. However, the issuances of a Certificate of Confidentiality helped to establish rapport with participants and promoted transparency. Lastly, our research that explored how fentanyl appears in the stimulant supply for use by people only intending to consume stimulants could be contextualized by future research exploring the perspectives of people who only use stimulants with and without opioids that incorporates community drug checking to explore how substance use practices are related to stimulant and opioid-involved overdose deaths.

5. Conclusion

This study characterized theorized pathways through which fentanyl may appear in the stimulant supply from the perspective of people who were incarcerated for distributing and/or manufacturing drugs. The belief that a systematic process of adulterating stimulants with fentanyl is underway in the US remains unsubstantiated based on our evidence. Instead, findings suggest that the rise in stimulant and opioid-involved overdose deaths may in part be the result of unintentional, spurious error resulting in fentanyl in or misattributed as stimulants. This conflation reflects the increasing unpredictability of an unregulated toxic drug supply in North America and underscores the need for creative harm reduction efforts to stabilize the drug supply and save lives.