Abstract
Background
We examined fentanyl use and outcomes among people who inject drugs (PWID) in the urban Pacific Northwest (PNW).
Methods
We conducted secondary analysis of data from the 2022 National HIV Behavioral Surveillance PWID cycle from Seattle, WA and Portland, OR. We calculated proportions of any fentanyl, injection fentanyl, non-injection fentanyl, and street pill use. Next, we compared characteristics, behaviors, and outcomes by any fentanyl use. Finally, we estimated adjusted relative risk (aRR) of overdose outcomes associated with fentanyl use and injecting less associated with type of street pill use.
Results
Among 830 PWID, 76% reported any fentanyl use, 47% specifically reported injection fentanyl use, 68% specifically reported non-injection fentanyl use, and 66% reported street pill use. Of those who used pills, 94% smoked them. Fentanyl use was crudely associated with age; race; housing; detainment; usual drug injected; injection frequency; syringe service programs, substance treatment, non-fatal overdose, and naloxone (p < 0.05). With adjustment, any fentanyl use (aRR = 3.20, 95% CI 1.58–6.47), injection fentanyl use (aRR = 3.51, 95% CI 1.63–7.58), and non-injection fentanyl use (aRR = 2.29, 95% CI 1.23–4.28) were associated with higher risk of overdose. Participants who smoked street pills were more likely to report injecting less because of pill use (aRR = 1.82, 95% CI 1.24–2.66).
Conclusions
Fentanyl use was highly prevalent among PWID in the urban PNW. Non-injection fentanyl use was most common and street pills were frequently administered by smoking which was associated with injecting less. All fentanyl use was associated with increased risk for overdose. There is ongoing need for access to naloxone for PWID who use fentanyl.
Keywords: Fentanyl, IDU, Naloxone, NIDU, Overdose, PWID
Background
Considered the “third wave” of the opioid epidemic, illicit fentanyl use is increasingly common among people who inject drugs (PWID) in the United States (US) [1]. Similar to heroin, and sometimes combined with heroin or other drugs, fentanyl is a synthetic opioid often distributed in powder form which can be injected, snorted, ingested or smoked [2]. Additionally, a growing proportion of fentanyl is in the form of street pills designed to look like prescription opioids [3]. Fentanyl is inherently more potent, faster acting, and generates a shorter high than heroin, lending itself to high risk for misuse, addiction, and opioid overdose [4]. Since the introduction of fentanyl into the US drug market, overdose cases have increased exponentially with fentanyl documented as a growing contributor [5]. Indeed, a recent report found almost two-thirds of drug overdose deaths from May 2020 to April 2021 in the US were attributed to illicitly manufactured fentanyl and fentanyl analogs [6].
Naloxone is an opioid antagonist that safely reverses the effects of opioid overdose and has been critical in reducing opioid-related morbidity and mortality associated with heroin and prescription opioids [7]. However, naloxone has demonstrated variable effectiveness at reducing risk of overdose due to fentanyl, with a frustrating level of failed attempts at resuscitation when compared to heroin overdose response [8]. The high potency and rapid onset of fentanyl may require quicker administration and additional doses of naloxone for consistent successful rescue [9].
Recent evidence indicates that the rise of fentanyl, in injection and non-injection form (e.g., smoked, snorted, or ingested) may also be influencing trends in injection frequency among PWID. While at least one study suggests PWID who use fentanyl inject at a greater frequency than those who do not use fentanyl (potentially due to fentanyl’s short duration of effect) [10], multiple studies indicate non-injection fentanyl use among PWID is associated with a decrease in injection drug use [11–13]. Though non-injection drug use has its own innate risks, a decrease in injection among PWID could be considered harm reduction as it relates to both overdose and infectious disease risk [14].
PWID in the Pacific Northwest (PNW, i.e., Washington and Oregon State) of the US may be uniquely affected by the fentanyl crisis. Analysis of the movement of illicit fentanyl through the US drug markets suggests it first emerged on the East Coast and has spread toward the West [15]. So while evaluation of the impact of fentanyl in Eastern US states like New York and Maryland has been possible since as early as 2014, a meaningful increase in fentanyl was only observed in Washington and Oregon in 2020 and 2021, respectively [16]. Therefore, to date, no studies have estimated the prevalence of fentanyl use among PWID in these areas nor described associated factors and outcomes [17].
The purpose of this study was to examine fentanyl use among PWID in the urban PNW. The objectives were to (1) quantify and describe fentanyl use and (2) explore the relationships between fentanyl use and key outcomes including overdose, naloxone ownership, and injection frequency.
Methods
Sample
We used data from the National HIV Behavioral Surveillance (NHBS) project, a program of the Centers for Disease Control and Prevention (CDC) that collects data in US metropolitan areas with high HIV prevalence [18]. In each area, NHBS teams collect cross-sectional data from three annually rotating populations considered to be at increased risk for HIV: men who have sex with men (MSM), low-income individuals at increased risk of HIV through heterosexual sex (HET), and PWID. For this analysis, we used data from the 2022 PWID cycle from Seattle, Washington and Portland, Oregon.
Participants in the 2022 PWID cycle were recruited from June through December 2022. Individuals were recruited through respondent driving sampling (RDS), a variation of chain-referral sampling designed to reach marginalized populations [19]. Recruitment began with a small convenience sample of initial participants, or “seeds.” These individuals then recruited new participants from their social networks, who subsequently recruited new participants from their social networks, and so on until the target sample size for each site (N = 500) was reached or the CDC-determined data collection period concluded, whichever came first.
Participants were eligible for the 2022 PWID cycle if they were at least 18 years old, resided in a qualifying county, were able to complete the survey in English or Spanish, and reported injecting drugs in the past 12 months. Eligible participants completed an anonymous computer-assisted survey with an off-site interviewer using video-conferencing software. The survey captured information about demographics, social, economic, and behavioral vulnerability to HIV, and access to HIV testing, care, and prevention. Participants were given Visa gift cards (Seattle) or cash (Portland) incentives for participation. Specifically, participants in Seattle were given $50 for the survey and $20 for referring peers; participants in Portland were given $40 for the survey and $10 for referring peers. Participants were also given the option of anonymous HIV and hepatitis C (HCV) testing, incentivized at $25 per test in Seattle and $20 per test in Portland.
Measures
Using supplemental questions to the CDC-mandated core NHBS survey, we operationalized four types of past 12 month fentanyl use (1) any fentanyl use (participant indicated that they used injection fentanyl, non-injection fentanyl, or street pills in the past 12 months), (2) any injection fentanyl use (participant specifically indicated that they injected fentanyl in the past 12 months), (3) any non-injection fentanyl use (participant specifically indicated that they used non-injection fentanyl in the past 12 months), and (4) any (presumed) fentanyl use via street pills. Street pills were defined as “pills from the street or online that look like Percocets, are often blue, have an M30 imprint, and may be called “blues,” “percs,” or “M30s.”
We next identified participant characteristics and injection drug use behaviors hypothesized to be associated with fentanyl use and our outcomes of interest based on existing literature. Participant characteristics included age [20, 21], gender [20, 22], race/ethnicity [23], past 12 months houselessness [24], past 12 months detention, jail, or imprisonment [25], and usual source of healthcare (“Is there a place you usually go when you are sick or you need advice about your health?”) [26], For injection drug use behaviors, we considered the drug that participants usually injected in the past 12 months [21, 27]. Participants could choose one option among heroin, methamphetamine, powder or crack cocaine, speedball [cocaine and heroin], or something else. Participants who answered “something else” specified with an open-ended response which drug they used. Based on frequency of report, responses of goofball [methamphetamine and heroin] and fentanyl or other drugs containing fentanyl were coded into their own categories. Other injection drug use behaviors included past 12 months injection frequency [10, 27, 28], any past 12 months syringe service program (SSP) use (to acquire sterile syringes) [29], and any past 12 months substance use treatment [30]. All participant characteristics and injection drug behaviors were derived from questions in the CDC-mandated core NHBS survey.
We then operationalized three outcomes associated with fentanyl use based on existing literature including opioid overdose [6], naloxone possession [8, 22], and changes in injection frequency [10, 12, 14, 31]. Opioid overdose was measured by a question in the NHBS core survey asked only of participants who reported past 12 months non-fentanyl opioid use: “In the past 12 months, did you overdose on heroin or painkillers? By overdose, I mean if you passed out, turned blue, or stopped breathing from using drugs.” Questions about naloxone possession were supplemental to the core survey and site-specific. Seattle participants were asked, “In the past 12 months, have you had a naloxone or Narcan kit?” and Portland participants were asked, “Do you currently have any naloxone or Narcan?” Finally, changes in injection frequency were assessed by both sites in a supplemental question specifically for participants who reported street pill use with the question, “In the past 12 months, did you inject less because you used these pills?” This question was specific to participants who reported street pill use because street pill use was hypothesized to be the most likely predictor of decreased injection among PWID in the PNW metropolitan areas.
Analysis
We first summarized participant characteristics and injection behaviors and outcomes for the whole sample and stratified by site.
To quantify fentanyl use, we then calculated the percentages of participants who reported different types of fentanyl use (any, injection, non-injection, and street pills) in the past 12 months. For individual types of fentanyl use, we also reported frequency of use, and for participants who reported street pill use, we reported how pills were used. Next, to describe fentanyl use, we stratified the total sample by any fentanyl use and compared proportions of participant characteristics, injection drug use behaviors, and overdose related outcomes, testing for differences across group with χ2 tests of independence.
Finally, to explore relationships between fentanyl use and outcomes of interest, we used generalized linear models with log-links, Poisson distribution and robust standard errors to estimate relative risk of each outcome associated with fentanyl use, adjusting for participant characteristics and behaviors hypothesized a priori to be associated with fentanyl use. For each model, we reported estimates of the adjusted relative risk (aRR) and 95% confidence intervals (95% CI) with statistical significance set a priori at p < 0.05 and assessed using Wald tests.
For the first set of multivariable models, we estimated the aRR of past 12 months overdose and naloxone ownership (current or past 12 months) associated with any (injection or non-injection) fentanyl use. In the second set of multivariable models, we included any injection fentanyl use, any non-injection fentanyl use, and an interaction term between any injection and any non-injection fentanyl use to allow us to estimate the individual and combined effects of any injection and any non-injection fentanyl use when compared to no fentanyl use on past 12 months overdose and naloxone ownership. Lastly, using a sub-sample of only participants who reported street pill use, we estimated the aRR of injecting less associated with different forms of street pill administration.
For all variables, skipped questions and responses of “Refuse to Answer” or “Don’t Know” were coded as missing. Participants who had missing data for all types of fentanyl use (injection, non-injection, and street pills) were dropped from the analytic sample. Multivariable models utilized complete case analysis and participants missing covariate or outcome data were dropped from relevant analyses. Analyses were conducted using Stata/IC v16 [32].
Portland and Seattle NHBS data collection and data sharing procedures were reviewed and approved by the Oregon Health Authority Science and Epidemiology Council’s Project Review Team and the Washington State IRB, respectively.
Results
A total of 835 participants were interviewed (500 from Seattle and 335 from Portland). Among these, 5 participants had missing data for all types of fentanyl use and were removed from the sample. Table 1 summarizes participants in the entire analytic sample (N = 830) and by site. Similar across sites, participants had a median age of 44 years old (IQR = 36–54 years old) and were most commonly cisgender men and white. Most participants reported experiencing houselessness in the past 12 months (62% in Seattle and 72% in Portland) and 28% reported being held in detention, jail or prison in the past 12 months. A high proportion of the total sample (87%) reported a usual source of healthcare.
Table 1.
Participant characteristics, injection drug use, and injection-drug use behaviors outcomes among people who inject drugs who participated in NHBS in two Pacific Northwest metropolitan areas, 2022 (N = 830)
| Seattle, WA (N = 497) | Portland, OR (N = 333) | Total (N = 830) | ||||
|---|---|---|---|---|---|---|
| Characteristics | N | (%) | N | (%) | N | (%) |
| Gendera | ||||||
| Woman | 182 | (37%) | 126 | (38%) | 308 | (37%) |
| Man | 312 | (63%) | 201 | (61%) | 513 | (62%) |
| Transgender | 3 | (1%) | 5 | (1%) | 8 | (1%) |
| Age | ||||||
| 18–29 | 28 | (6%) | 26 | (8%) | 54 | (7%) |
| 30–39 | 115 | (23%) | 118 | (35%) | 233 | (28%) |
| 40–49 | 145 | (29%) | 95 | (29%) | 240 | (29%) |
| 50–60 | 135 | (27%) | 55 | (17%) | 190 | (23%) |
| 60 + | 74 | (15%) | 39 | (12%) | 113 | (14%) |
| Race/Ethnicitya,b | ||||||
| American Indian or Alaska native | 76 | (15%) | 39 | (12%) | 115 | (14%) |
| Asian | 12 | (2%) | 6 | (2%) | 18 | (2%) |
| Black or African American | 115 | (23%) | 39 | (12%) | 154 | (19%) |
| Native Hawaiian or Pacific Islander | 32 | (6%) | 5 | (2%) | 37 | (4%) |
| White | 339 | (69%) | 284 | (86%) | 623 | (76%) |
| Hispanic | 53 | (11%) | 24 | (7%) | 77 | (9%) |
| Experienced houselessnessc | 308 | (62%) | 240 | (72%) | 548 | (66%) |
| Detained/Jailed/Imprisonedc | 146 | (29%) | 83 | (25%) | 229 | (28%) |
| Has a usual source of healthcare | 435 | (88%) | 285 | (86%) | 720 | (87%) |
| Injection drug use behaviorsc | ||||||
| Usual drug injected | ||||||
| Heroin | 212 | (43%) | 173 | (52%) | 385 | (46%) |
| Methamphetamine | 140 | (28%) | 123 | (37%) | 263 | (32%) |
| Powder or crack cocaine | 12 | (2%) | 4 | (1%) | 16 | (2%) |
| Speedball (cocaine + heroin) | 46 | (9%) | 8 | (2%) | 54 | (7%) |
| Goofball (methamphetamine + heroin) | 57 | (11%) | 14 | (4%) | 71 | (9%) |
| Fentanyl or drugs containing fentanyl | 17 | (3%) | 8 | (2%) | 25 | (3%) |
| Other drugs or combination of drugs | 7 | (1%) | 0 | (0%) | 7 | (1%) |
| Injection frequencya | ||||||
| More than once a day | 283 | (57%) | 219 | (66%) | 505 | (60%) |
| Once a day | 59 | (12%) | 37 | (11%) | 96 | (12%) |
| More than once a week | 75 | (15%) | 36 | (11%) | 113 | (14%) |
| Once a week or less | 77 | (16%) | 41 | (12%) | 118 | (14%) |
| Syringe service program usea | 359 | (73%) | 246 | (74%) | 605 | (73%) |
| Substance use treatment | 151 | (30%) | 105 | (32%) | 256 | (31%) |
| Overdose outcomes | ||||||
| Experienced overdosec,d | 121 | (27%) | 73 | (26%) | 194 | (23%) |
| Naloxone ownershipe | 406 | (82%) | 237 | (71%) | 643 | (77%) |
aMissing data, bAlone or in combination; not mutually exclusive, cIn past 12 months, dLimited to participants who reported non-fentanyl opioid use (N = 721), eRefers to got naloxone in past 12 months for Seattle and currently has naloxone for Portland
Patterns of usual drug injected in the past 12 months were also similar across Seattle and Portland; the largest proportion reported heroin (43 and 52%, respectively) and the next largest proportion reported methamphetamine (28 and 37% respectively). Fentanyl was reported as the usual drug injected by 3% of the Seattle sample and 2% of the Portland sample. Fifty-seven percent of the Seattle sample and 66% of the Portland sample reported injecting more than once a day in the past 12 months. Seventy-three percent of the Seattle sample and 74% of the Portland sample reported acquiring sterile syringes at SSPs in the past 12 months. A little under a third of participants in each site reported substance treatment in the past 12 months. Consistent across sites, among participants who reported non-fentanyl opioid use (N = 721), about one quarter reported experiencing an opioid overdose in the past 12 months. Eighty-two percent of the Seattle sample reported having naloxone in the past 12 months and 71% of the Portland sample reported currently having naloxone.
Table 2 shows fentanyl use among the entire sample. Over three-fourths of the sample reported any fentanyl use in the past 12 months. Fentanyl was less commonly injected (47%) when compared to any non-injection or street pill use (68% and 66%, respectively). Frequency of non-injection fentanyl use was highly correlated with frequency of street pill use (post hoc Spearman’s correlation = 0.79, p < 0.001); 42% reported using non-injection fentanyl once a day or more and 41% reported using street pills once a day or more. Among the 545 participants who reported using street pills, 94% reported smoking them with much smaller percentages reporting injecting (12%), snorting (13%), ingesting (11%), or other forms of administering the pills (1%).
Table 2.
Past 12 months fentanyl use among people who inject drugs who participated in NHBS in two Pacific Northwest metropolitan areas, 2022
| Na | % of total Nb | |
|---|---|---|
| Fentanyl use | (N = 830) | |
| No fentanyl use | 199 | (24%) |
| Any fentanyl use | 631 | (76%) |
| Injection fentanyl use | (N = 805) | |
| No injection fentanyl use | 423 | (53%) |
| Any injection fentanyl use | 382 | (47%) |
| More than once a day | 132 | (16%) |
| Once a day or more | 31 | (4%) |
| More than once a week | 82 | (10%) |
| Once a week or less | 137 | (17%) |
| Non-injection fentanyl use | (N = 819) | |
| No non-injection fentanyl use | 263 | (32%) |
| Any non-injection fentanyl use | 556 | (68%) |
| More than once a day | 267 | (33%) |
| Once a day | 70 | (9%) |
| More than once a week | 94 | (11%) |
| Once a week or less | 125 | (15%) |
| Street pill use | (N = 828) | |
| No street pill use | 283 | (34%) |
| Any street pill use | 545 | (66%) |
| More than once a day | 279 | (34%) |
| Once a day | 62 | (7%) |
| More than once a week | 81 | (10%) |
| Once a week or less | 123 | (15%) |
| How street pills were usedc | (N = 545) | |
| Injected | 67 | (12%) |
| Smoked | 511 | (94%) |
| Snorted | 72 | (13%) |
| Swallowed/Ingested | 61 | (11%) |
| Other | 3 | (1%) |
aSkipped questions and responses of “Refuse to Answer” or “Don’t Know” were coded as missing for individual types of fentanyl use bItalicized percentages of reported frequency of use sum to the preceding total percent of any use. cNot mutually exclusive categories; column totals will not equal 100% for each type of street pill use.
In bivariate analysis (Table 3), compared to participants who reported no fentanyl use in the past 12 months, participants who reported any fentanyl use had a younger age distribution, lower representation of Black or African American participants, higher representation of white participants, and higher proportions of houselessness and detention, jail, or imprisonment (p < 0.001). Participants who reported any fentanyl use also appeared to have different patterns of usual drug injected (p < 0.001). Specifically, participants who reported any fentanyl use had higher proportions of participants who usually injected heroin or goofballs (methamphetamine and heroin), and lower proportions of participants who usually injected just methamphetamine when compared to participants who reported no fentanyl use. Injection frequency also differed by fentanyl use (p < 0.001); participants who reported fentanyl use were more likely to inject more than once a day when compared to those who did not (66% vs 45%, respectively). Past 12 months SSP use and substance use treatment were both significantly more common among participants who used fentanyl (p < 0.01 and p = 0.05, respectively). Participants who reported any past 12 months fentanyl use were also more likely to report having experienced an overdose and past 12 months or current naloxone ownership (p < 0.001).
Table 3.
Participant characteristics, injection drug use, and injection-drug use behaviors outcomes, stratified by any fentanyl use, among people who inject drugs who participated in NHBS in two Pacific Northwest metropolitan areas, 2022 (N = 830)
| No fentanyl (N = 199) | Any fentanyl (N = 631) | ||||
|---|---|---|---|---|---|
| Characteristics | N | (%) | N | (%) | p-value |
| Gendera | 0.38 | ||||
| Woman | 80 | (40%) | 228 | (36%) | |
| Man | 116 | (58%) | 397 | (63%) | |
| Transgender | 3 | (2%) | 5 | (1%) | |
| Age | < 0.001 | ||||
| 18–29 | 3 | (1%) | 52 | (8%) | |
| 30–39 | 24 | (12%) | 209 | (33%) | |
| 40–49 | 50 | (25%) | 190 | (30%) | |
| 50–60 | 63 | (32%) | 127 | (20%) | |
| 60 + | 60 | (30%) | 53 | (8%) | |
| Race/ethnicitya | |||||
| American Indian or Alaska native | 27 | (14%) | 88 | (14%) | 0.89 |
| Asian | 2 | (1%) | 16 | (3%) | 0.20 |
| Black or African American | 67 | (34%) | 87 | (14%) | < 0.001 |
| Native Hawaiian or Pacific Islander | 10 | (5%) | 27 | (4%) | 0.66 |
| White | 123 | (62%) | 500 | (79%) | < 0.001 |
| Hispanic | 19 | (10%) | 58 | (9%) | 0.89 |
| Experienced houselessnessc | 96 | (48%) | 452 | (72%) | < 0.001 |
| Detained/Jailed/Imprisonedc | 24 | (12%) | 205 | (32%) | < 0.001 |
| Has a usual source of healthcare | 179 | (90%) | 541 | (86%) | 0.13 |
| Injection drug usec | |||||
| Usual drug injected | < 0.001 | ||||
| Heroin | 72 | (36%) | 313 | (50%) | |
| Methamphetamine | 99 | (50%) | 164 | (26%) | |
| Powder or crack cocaine | 9 | (5%) | 7 | (1%) | |
| Speedball (cocaine + heroin) | 13 | (7%) | 41 | (7%) | |
| Goofball (methamphetamine + heroin) | 5 | (3%) | 66 | (10%) | |
| Fentanyl or drugs containing fentanyl | 0 | (0%) | 25 | (4%) | |
| Other drugs or combination of drugs | 0 | (0%) | 7 | (1%) | |
| Injection frequencya | < 0.001 | ||||
| More than once a day | 89 | (45%) | 413 | (66%) | |
| Once a day | 38 | (19%) | 58 | (9%) | |
| More than once a week | 36 | (18%) | 75 | (12%) | |
| Once a week or less | 36 | (18%) | 82 | (13%) | |
| Syringe service program usea | 125 | (63%) | 480 | (76%) | < 0.01 |
| Substance use treatment | 50 | (25%) | 206 | (33%) | 0.05 |
| Overdose outcomes | |||||
| Experienced overdoseb,c | 8 | (7%) | 186 | (31%) | < 0.001 |
| Naloxone ownershipd | 124 | (62%) | 519 | (82%) | < 0.001 |
p value generated from Χ2
aMissing data, bAlone or in combination; not mutually exclusive, cIn past 12 months, dLimited to participants who reported non-fentanyl opioid use (N = 721), dRefers to got naloxone in past 12 months for Seattle and currently has naloxone for Portland
Table 4 shows results of the multivariable analyses. After adjusting for potential confounders, compared to those who had no fentanyl use, participants who used any fentanyl were approximately three times more likely to have experienced an overdose (aRR = 3.20, 95% CI 1.58, 6.47), but not more likely to have naloxone (aRR = 1.11, 95% CI 0.99, 1.23). When modeling the individual and interacting effects of injection and non-injection fentanyl use, all combinations were associated with an increased risk of overdose. When compared to those who reported no fentanyl use, participants who reported only injection fentanyl use had the highest risk of overdose (aRR = 3.51, 95% CI 1.63, 7.58), followed by those who reported injection and non-injection fentanyl use (aRR = 2.75, 95% CI 1.50, 5.03), and those who reported only non-injection fentanyl use (aRR = 2.29, 95% CI 1.23, 4.28). Naloxone ownership was not associated with only injection fentanyl use (aRR = 0.99, 95% CI 0.78, 1.26) or with only non-injection fentanyl use (aRR = 1.10, 95% CI 0.98, 1.24), but was associated with injection and non-injection fentanyl use (aRR = 1.13, 95% CI 1.01, 1.27), all when compared to no fentanyl use.
Table 4.
Adjusted relative risk of overdose outcomes associated with any, injection and/or non-injection fentanyl use, compared to no fentanyl use, among people who inject drugs who participated in NHBS in two Pacific Northwest metropolitan areas, 2022
| Experienced overdosea (N = 716) | Naloxone ownership (N = 824) | |||
|---|---|---|---|---|
| aRR | 95% CI | aRR | 95% CI | |
| 4a. Any fentanyl vs noneb | ||||
| Any fentanyl use | 3.20 | (1.58, 6.47) | 1.11 | (0.99, 1.23) |
| 4b. Injection, non-injection, or both vs nonec | ||||
| Only injection fentanyl use | 3.51 | (1.63, 7.58) | 0.99 | (0.78, 1.26) |
| Only non-injection fentanyl use | 2.29 | (1.23, 4.28) | 1.10 | (0.98, 1.24) |
| Injection and non-injection fentanyl use | 2.75 | (1.50, 5.03) | 1.13 | (1.01, 1.27) |
| Injection fentanyl use by non-injection fentanyl use interaction term p-value | 0.01 | 0.75 |
Bold-p < 0.05
aLimited to participants who reported non-fentanyl opioid use, bUsing 1 model for each outcome, cUsing 1 model for each outcome, each including an interaction term injection and non-injection fentanyl use. All models adjusted for city, gender, age, race/ethnicity, past 12 months houselessness, past 12 months detention/jail/imprisonment, usual source of healthcare, usual drug injected, injection frequency, past 12 months SSP use, past 12 months substance use treatment
Finally, among participants who reported street pill use, in adjusted analysis, when compared to participants who did not smoke pills, participants who reported smoking pills were almost two times more likely to report injecting less because of pill use (aRR = 1.82, 95% CI 1.24, 2.66). The associations between injecting, snorting, ingesting, and other forms of administering street pills and injecting less were all statistically nonsignificant (Table 5).
Table 5.
Adjusted relative risks of injecting less associated with type of drug administration, compared to not using that type of administration, among people who used fentanyl street pills in two Pacific Northwest metropolitan areas, 2022 (N = 538)a
| Reported injecting less because of pill use | ||
|---|---|---|
| Drug administration for street pills | aRR | 95% CI |
| Smoked | 1.82 | (1.24, 2.66) |
| Snorted | 1.05 | (0.87, 1.26) |
| Swallowed/ingested | 1.01 | (0.81, 1.26) |
| Other | 1.39 | (0.86, 2.24) |
Bold-p < 0.05
aUsing 1 model including individual dummy variables for each type of fentanyl drug administration, adjusted for city, gender, age, race/ethnicity, past 12 months houselessness, past 12 months detention/jail/imprisonment, usual source of healthcare, usual drug injected, injection frequency, past 12 months SSP use, past 12 months substance use treatment, and injecting street pills
Discussion
In a sample of PWID from two large PNW metropolitan areas in 2022, fentanyl use was highly prevalent. Fentanyl was injected by some PWID, but non-injection fentanyl and/or street pill use were more commonly reported, and street pills were most frequently administered by smoking. In bivariate analysis, fentanyl use was associated with participant characteristics and injection drug use behaviors. In adjusted analysis, any, injection, non-injection, and both injection and non-injection fentanyl use were all associated with meaningfully higher risk of overdose, but not necessarily associated with naloxone ownership. Finally, while any fentanyl use was crudely associated with higher injection frequency, in an adjusted analysis, participants who smoked street pills were significantly more likely to report injecting less because of their pill use than participants who used street pills in other ways.
Our finding that fentanyl use was common among PWID in the urban PNW is consistent with recent evidence that fentanyl has quickly and effectively infiltrated the drug markets of the Western US [15, 16]. Furthermore, our findings corroborate early qualitative work out of Central Oregon and a toxicology study in the Seattle area indicating fentanyl use is specifically impacting PWID in the PNW [33, 34]. Indeed, our data from this 2022 PWID cycle suggest that fentanyl use among PWID has rapidly increased in just a year. Using a Washington State SSP Health Survey report as a proxy for PNW fentanyl use in 2021, the report showed under half (42%) of SSP users reported fentanyl use in the past 3 months, much lower than the 76% of our PWID sample who reported fentanyl use in the past 12 months. This type of precipitous growth mirrors what was documented in the Eastern US just a few years prior [35, 36], and demonstrates the speed at which fentanyl can inundate a new market.
The high level of fentanyl use in our sample should be at least partially attributed to fentanyl’s highly addictive nature [37]. Multiple studies have documented an explicit preference for fentanyl by PWID [20, 38], and in a sample of PWID in Vancouver, Canada, many respondents indicated that fentanyl gives a “better” high than other opioids [21]. In addition, research suggests that fentanyl may be particularly easy to manufacture and distribute, increasingly in the form of counterfeit prescription pills [2, 3]. Two thirds of our sample reported using street pills in the past 12 months. This coincides with a recent study of online pill sales showing a large volume of counterfeit oxycodone pills offered at lower prices than pharmaceutical grade pills and targeted at US and Canadian customers [39], and anecdotal evidence in Washington State where local experts believe there are high volumes of street pill sales via the internet and social media apps [40].
Unlike some previous studies of fentanyl use among PWID [41], we found many PWID using fentanyl were not injecting it. In fact, non-injection fentanyl use seemed to be preferred, and among participants who reported street pills use, almost all reported smoking pills (compared to very small percentages of participants who reported injecting or administering pills in other ways). Furthermore, participants who reported smoking pills were significantly more likely to report injecting less because of using pills. These results are similar to findings from longitudinal studies of PWID in San Francisco and San Diego [12, 31] and are consistent with national trends, all showing decreases in injection and increases in smoking or inhalation of fentanyl among people who use opioids [13].
Interestingly, participants in our sample who used fentanyl in the past 12 months were more likely to have higher injection frequency when compared to participants who did not. This type of correlation has been reported in previous studies [10, 42], but is not necessarily evidence that fentanyl use leads to higher frequency injection. Instead, given our finding that smoking pills is explicitly associated with injecting less, it may be more likely that people who have a high injection frequency at baseline are choosing to smoke fentanyl (or administer fentanyl in other non-injecting ways) because they are trying to reduce risks associated with injecting. This hypothesis aligns with qualitative work out of Canada documenting that PWID are indeed transitioning to smoking fentanyl for harm reduction purposes [11, 43]. Moreover, PWID in San Francisco have reported beliefs that switching from injecting opioids to smoking fentanyl may reduce their risk of blood born illness transmission, abscesses and other skin infections, stigma associated with injecting, and most prominently, overdose [14, 31].
In our sample, any modality of fentanyl use was associated with significant increased risk for overdose. However, there did seem to be a slightly lower risk among people who only reported non-injection fentanyl use, providing some empirical evidence to support the idea that switching to non-injection fentanyl use is a strategy for reducing overdose risk. Still, according to a recent national study, in 2022, smoking was the most commonly documented route of use in overdose death [44]. As such, strategies to prevent overdose among people who use fentanyl, even those who do not inject it, should continue to be prioritized [8].
Encouragingly, a significantly higher proportion of participants in our sample who used fentanyl reported naloxone ownership compared to participants who had no fentanyl use. This is similar to findings from the 2018 NHBS-PWID cycle in New York City which found participants who reported recent known or possible fentanyl use were more likely to own naloxone [45]. It is also consistent with findings from qualitative work out of rural Oregon that showed most participants were aware of the influx of fentanyl and expressed anxiety regarding the associated increased risk for overdose [33].
Notably, some research indicates previous standards of naloxone ownership and use may be insufficient for preventing fentanyl overdose [46]. Specifically, the amount of naloxone and the window of time for administration required to reverse a fentanyl overdose may be greater and shorter, respectively, when compared to heroin [9]. Simultaneously, the intense symptoms of severe opioid withdrawal associated with rapid fentanyl overdose reversal also require consideration [47]. PWID who use fentanyl may benefit from additional naloxone distribution along with fentanyl-specific overdose training and safe spaces where immediate and tolerable administration of naloxone is consistently possible. A 2018 ethnographic study of a safe consumption site in Vancouver, Canada observed a 100% success rate of fentanyl overdose reversal by naloxone administration [48]. This was achieved through community trainings, vigilant monitoring, immediate response to potential overdose, and multiple doses of naloxone distributed and administered.
Historically, SSPs have played a critical role in overdose education and naloxone distribution (OEND) for PWID. A 2022 multi-site US study found most SSPs had substantially increased naloxone distribution and some had adapted overdose education in response to recent fentanyl overdose surges [49], indicating SSPs will continue to be a key resource in the third wave of the opioid epidemic. Indeed, most PWID in our sample reported accessing sterile syringes from SSPs in the past 12 months, and using SSPs was more common among people who used fentanyl. However, if PWID who use fentanyl are injecting less, or eventually not injecting at all, it will be important for SSPs to broaden their outreach and expand their scope of services to meet the needs of people who use drugs (PWUD) but may not inject them. For example, in the 2021 Washington State SSP survey, 72% of respondents indicated that they would like safe drug smoking supplies [50]. Unfortunately, at the time of publication, distribution of safe smoking supplies is restricted in many states, including Oregon. Still, SSPs are well positioned to provide other auxiliary supplies and services to PWUD who are using fentanyl including fentanyl testing strips, medication for opioid use disorder (MOUD), and linkage to substance abuse treatment and other support services.
Limitations
This analysis was subject to limitations. First, data were collected cross-sectionally and typically referenced the past 12 months for reported behaviors. As such, associations cannot generally be interpreted as causal. One exception is our question explicitly asking if participants injected less because they used street pills. Still, we asked that question only to participants who reported street pill use and assumed street pills contained fentanyl. Future research may benefit from asking participants about injecting less because of fentanyl use. Next, the NHBS core survey only asked about overdose if the participant reported using non-fentanyl opioids (heroin, heroin-containing polysubstances, and/or painkillers) so we were unable to assess experiences of overdose among people who used fentanyl but no other opioids. Future analyses would benefit from inclusion of an overdose question asked to all participants. This analysis was also limited by the difference in naloxone variables across sites. Fortunately, similar proportions within each site sample indicated they were both reasonable proxies for overdose harm reduction via naloxone. Next, past studies have shown fentanyl use associated with injection risk behaviors (e.g., sharing syringes) [10, 28, 41]. Based on local observations and anecdotal evidence, we focused on just a decrease in injecting. Our study sample may also be limited in generalizability. Both sites used RDS methods; however, it is unlikely that either site met all assumptions needed for inference to the underlying populations (e.g., sufficient number of recruitment waves, accurate report of network size, and non-preferential recruitment) [51]. Furthermore, caution should be taken when applying findings to populations in other regions of the US. Lastly, our sample only included people who injected in the past 12 months. Based on our analysis, it is possible that some former PWID stopped injecting just before they would have been eligible for our study. It may be helpful for future studies to also collect data from people who injected within the past few years.
Conclusions
In summary, fentanyl use was highly prevalent among PWID in two large PNW metropolitan areas in 2022. Non-injection fentanyl was more commonly reported than injection fentanyl use, and street pills were most frequently administered by smoking. Though smoking street pills was associated with a decrease in injection, our study indicated fentanyl use (including only non-injection fentanyl use) was associated with a substantial increased risk for overdose. Naloxone ownership was high in our sample, but potentially insufficient to counter the risk for overdose associated with fentanyl use. There is a continued need to ensure low-barrier access to naloxone with a particular emphasis on people who use fentanyl and their social networks. Fentanyl-specific overdose education, safe consumptions sites, and SSPs with expanded service provision can all contribute toward ensuring PWID and former-PWID who use fentanyl can reduce their risk for overdose and death.
Acknowledgements
The authors would like to thank the 2022 NHBS project staff and participants in Seattle and Portland for their invaluable contributions to research and public health.
Abbreviations
- PWID
People who inject drugs
- PNW
Pacific Northwest
- aRR
Adjusted relative risk
- US
United States
- NHBS
National HIV behavioral surveillance
- CDC
Centers for disease control and prevention
- MSM
Men who have sex with men
- HET
Low-income individuals at increased risk of HIV through heterosexual sex
- SSP
Syringe service program
- CI
Confidence intervals
- OEND
Overdose education and naloxone distribution
- PWUD
People who use drugs
- MOUD
Medication for opioid use disorder
Author contributions
TM and SG were principal investigators for the study sites in 2022 and lead the sites’ investigations and provided study supervision. LL and CM helped administer the project and coordinated data collection and data management. LL analyzed and interpreted the data and drafted the manuscript. All authors provided feedback on early drafts of the manuscript and all authors reviewed, edited, and approved the final manuscript.
Funding
This work was supported by the Centers for Disease Control and Prevention National HIV Behavioral Surveillance Project (5 NU62PS924776-03, 5 NU62PS924762-03).
Data availability
The datasets used and/or analyzed during the current study may be available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
All participants provided verbal, documented, informed consent for study activities. Portland and Seattle NHBS data collection and data sharing procedures were reviewed and approved by the Oregon Health Authority Science and Epidemiology Council’s Project Review Team and the Washington State IRB (WSIRB), respectively.
Consent for publication
Not applicable.
Competing interests
SG reports a relationship with Cepheid that includes funding grants. LS reports financial support was provided by Centers for Disease Control and Prevention. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Ciccarone D. The triple wave epidemic: supply and demand drivers of the US opioid overdose crisis. Int J Drug Policy. 2019;71:183–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Pergolizzi J, Magnusson P, LeQuang JAK, Breve F. Illicitly manufactured fentanyl entering the United States. Cureus. 2021;13(8):e17496. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Palamar JJ, Ciccarone D, Rutherford C, Keyes KM, Carr TH, Cottler LB. Trends in seizures of powders and pills containing illicit fentanyl in the United States, 2018 through 2021. Drug Alcohol Depend. 2022;234:109398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Suzuki J, El-Haddad S. A review: fentanyl and non-pharmaceutical fentanyls. Drug Alcohol Depend. 2017;171:107–16. [DOI] [PubMed] [Google Scholar]
- 5.Jones CM, Einstein EB, Compton WM. Changes in synthetic opioid involvement in drug overdose deaths in the United States, 2010–2016. JAMA. 2018;319(17):1819–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.O’Donnell J, Tanz LJ, Gladden RM, Davis NL, Bitting J. Trends in and characteristics of drug overdose deaths involving illicitly manufactured fentanyls—United States, 2019–2020. MMWR Morb Mortal Wkly Rep. 2021;70(50):1740–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kim D, Irwin KS, Khoshnood K. Expanded access to naloxone: options for critical response to the epidemic of opioid overdose mortality. Am J Public Health. 2009;99(3):402–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fairbairn N, Coffin PO, Walley AY. Naloxone for heroin, prescription opioid, and illicitly made fentanyl overdoses: challenges and innovations responding to a dynamic epidemic. Int J Drug Policy. 2017;46:172–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Somerville NJ, O’Donnell J, Gladden RM, Zibbell JE, Green TC, Younkin M, et al. Characteristics of fentanyl overdose—Massachusetts, 2014–2016. MMWR Morb Mortal Wkly Rep. 2017;66(14):382–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lambdin BH, Bluthenthal RN, Zibbell JE, Wenger L, Simpson K, Kral AH. Associations between perceived illicit fentanyl use and infectious disease risks among people who inject drugs. Int J Drug Policy. 2019;74:299–304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kamal A, Ferguson M, Xavier JC, Liu L, Graham B, Lock K, et al. Smoking identified as preferred mode of opioid safe supply use; investigating correlates of smoking preference through a 2021 cross-sectional study in British Columbia. Subst Abuse Treat Prev Policy. 2023;18(1):27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Eger WH, Abramovitz D, Bazzi AR, Bórquez A, Vera CF, Harvey-Vera A, et al. Changes in injecting versus smoking heroin, fentanyl, and methamphetamine among people who inject drugs in San Diego, California, 2020-2023. Drug Alcohol Depend. 2024;259:111318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Karandinos G, Unick J, Ondocsin J, Holm N, Mars S, Montero F, et al. Decrease in injection and rise in smoking and snorting of heroin and synthetic opioids, 2000-2021. Drug Alcohol Depend. 2024;263:111419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ciccarone D, Holm N, Ondocsin J, Schlosser A, Fessel J, Cowan A, et al. Innovation and adaptation: the rise of a fentanyl smoking culture in San Francisco. PLoS ONE. 2024;19(5):e0303403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Shover CL, Falasinnu TO, Dwyer CL, Santos NB, Cunningham NJ, Freedman RB, et al. Steep increases in fentanyl-related mortality west of the Mississippi River: recent evidence from county and state surveillance. Drug Alcohol Depend. 2020;216:108314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zoorob MJ, Park JN, Kral AH, Lambdin BH, Del Pozo B. Drug decriminalization, fentanyl, and fatal overdoses in Oregon. JAMA Netw Open. 2024;7(9):e2431612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Tsang VWL, Wong JSH, Westenberg JN, Ramadhan NH, Fadakar H, Nikoo M, et al. Systematic review on intentional non-medical fentanyl use among people who use drugs. Front Psychiatry. 2024;15:1347678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Gallagher KM, Sullivan PS, Lansky A, Onorato IM. Behavioral surveillance among people at risk for HIV infection in the U.S.: the national HIV behavioral surveillance system. Public Health Rep. 2007;122(Suppl 1):32–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Heckathorn DD. Respondent-driven sampling: a new approach to the study of hidden populations. Soc Probl. 1997;44(2):174–99. [Google Scholar]
- 20.Mazhnaya A, O’Rourke A, White RH, Park JN, Kilkenny ME, Sherman SG, et al. Fentanyl preference among people who inject drugs in West Virginia. Subst Use Misuse. 2020;55(11):1774–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ickowicz S, Kerr T, Grant C, Milloy MJ, Wood E, Hayashi K. Increasing preference for fentanyl among a cohort of people who use opioids in Vancouver, Canada, 2017–2018. Subst Abuse. 2022;43(1):458–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Jones AA, Schneider KE, Mahlobo CT, Maggs JL, Dayton L, Tobin KE, et al. Fentanyl overdose concerns among people who inject drugs: the role of sex, racial minority status, and overdose prevention efforts. Psychol Addict Behav. 2023;37(2):191–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Cano M, Mendoza N, Ignacio M, Rahman A, Daniulaityte R. Overdose deaths involving synthetic opioids: racial/ethnic and educational disparities in the eastern and western US. Drug Alcohol Depend. 2023;251:110955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Milaney K, Passi J, Zaretsky L, Liu T, O’Gorman CM, Hill L, et al. Drug use, homelessness and health: responding to the opioid overdose crisis with housing and harm reduction services. Harm Reduct J. 2021;18(1):92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kaplowitz E, Truong AQ, Macmadu A, Peterson M, Brinkley-Rubinstein L, Potter N, et al. Fentanyl-related overdose during incarceration: a comprehensive review. Health Justice. 2021;9(1):13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Sahebi-Fakhrabad A, Sadeghi AH, Kemahlioglu-Ziya E, Handfield R. Exploring opioid prescription patterns and overdose rates in South Carolina (2017-2021): insights into rising deaths in high-risk areas. Healthcare. 2024. 10.3390/healthcare12131268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Chandra DK, Altice FL, Copenhaver MM, Zhou X, Didomizio E, Shrestha R. Purposeful fentanyl use and associated factors among opioid-dependent people who inject drugs. Subst Use Misuse. 2021;56(7):979–87. [DOI] [PubMed] [Google Scholar]
- 28.Park JN, Owczarzak J, Urquhart G, Morris M, Weicker NP, Rouhani S, et al. HIV risk among urban and suburban people who inject drugs: elevated risk among fentanyl and cocaine injectors in Maryland. AIDS Behav. 2022;26(1):277–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Packham A. Syringe exchange programs and harm reduction: new evidence in the wake of the opioid epidemic. J Public Econ. 2022;215:104733. [Google Scholar]
- 30.Gryczynski J, Nichols H, Schwartz RP, Mitchell SG, Hill P, Wireman K. Fentanyl exposure and preferences among individuals starting treatment for opioid use disorder. Drug Alcohol Depend. 2019;204:107515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Kral AH, Lambdin BH, Browne EN, Wenger LD, Bluthenthal RN, Zibbell JE, et al. Transition from injecting opioids to smoking fentanyl in San Francisco, California. Drug Alcohol Depend. 2021;227:109003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.StataCorp. (2019) Stata statistical software: release 16. College Station, TX: StataCorp LLC
- 33.Shin SS, LaForge K, Stack E, Pope J, Leichtling G, Larsen JE, et al. It wasn’t here, and now it is. It’s everywhere": fentanyl’s rising presence in Oregon’s drug supply. Harm Reduct J. 2022;19(1):76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Reinhart AA, Simpson C, Yarid N, Haruff R, Baird GS. Investigation of fentanyl usage in the Seattle area. J Appl Lab Med. 2022;7(5):1164–8. [DOI] [PubMed] [Google Scholar]
- 35.Colon-Berezin C, Nolan ML, Blachman-Forshay J, Paone D. Overdose deaths involving fentanyl and fentanyl Analogs—New York City, 2000–2017. MMWR Morb Mortal Wkly Rep. 2019;68(2):37–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Lu H, Crawford FW, Gonsalves GS, Grau LE. Geographic and temporal trends in fentanyl-detected deaths in Connecticut, 2009–2019. Ann Epidemiol. 2023;79:32–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Comer SD, Cahill CM. Fentanyl: receptor pharmacology, abuse potential, and implications for treatment. Neurosci Biobehav Rev. 2019;106:49–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Hochstatter KR, Terplan M, Mitchell SG, Schwartz RP, Dusek K, Wireman K, et al. Characteristics and correlates of fentanyl preferences among people with opioid use disorder. Drug Alcohol Depend. 2022;240:109630. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Lamy FR, Daniulaityte R, Dudley S Jr. Pressed OXY M30 Pills, Great Press, Potent, Fast Shipping!!!": availability of counterfeit and pharmaceutical oxycodone pills on one major cryptomarket. J Psychoactive Drugs. 2024;56(1):1–7. [DOI] [PubMed] [Google Scholar]
- 40.UW Medicine Newsroom. 2021. [cited 2024]. Available from: https://newsroom.uw.edu/blog/qa-fentanyls-growth-among-overdoses-stunning.
- 41.Kenny KS, Kolla G, Greig S, Bannerman M, Phillips D, Altenberg J, et al. Association of illicit fentanyl use with injection risk practices among people who inject drugs. AIDS Behav. 2023;27(6):1757–65. [DOI] [PubMed] [Google Scholar]
- 42.Duhart Clarke SE, Kral AH, Zibbell JE. Consuming illicit opioids during a drug overdose epidemic: illicit fentanyls, drug discernment, and the radical transformation of the illicit opioid market. Int J Drug Policy. 2022;99:103467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Bardwell G, Austin T, Maher L, Boyd J. Hoots and harm reduction: a qualitative study identifying gaps in overdose prevention among women who smoke drugs. Harm Reduct J. 2021;18(1):29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Tanz LJ, Gladden RM, Dinwiddie AT, Miller KD, Broz D, Spector E, et al. Routes of drug use among drug overdose deaths—United States, 2020–2022. MMWR Morb Mortal Wkly Rep. 2024;73(6):124–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Rivera AV, Nolan ML, Paone D, Carrillo SA, Braunstein SL. Gaps in naloxone ownership among people who inject drugs during the fentanyl wave of the opioid overdose epidemic in New York City, 2018. Subst Abuse. 2022;43(1):1172–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Coffin PO, Maya S, Kahn JG. Modeling of overdose and naloxone distribution in the setting of fentanyl compared to heroin. Drug Alcohol Depend. 2022;236:109478. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Lai JT, Goldfine CE, Chapman BP, Taylor MM, Rosen RK, Carreiro SP, et al. Nobody wants to be narcan’d: a pilot qualitative analysis of drug users’ perspectives on naloxone. West J Emerg Med. 2021;22(2):339–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Mayer S, Boyd J, Collins A, Kennedy MC, Fairbairn N, McNeil R. Characterizing fentanyl-related overdoses and implications for overdose response: findings from a rapid ethnographic study in Vancouver, Canada. Drug Alcohol Depend. 2018;193:69–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Frost MC, Austin EJ, Corcorran MA, Briggs ES, Behrends CN, Juarez AM, et al. Responding to a surge in overdose deaths: perspectives from US syringe services programs. Harm Reduct J. 2022;19(1):79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Kingston S, Newman A, Banta-Green C, Glick S. Results from the 2021 WA State Syringe Service Program Health Survey. Addictions, Drug & Alcohol Institute, Department of Psychiatry & Behavioral Sciences, University of Washington: Seattle, WA. 2022.
- 51.Gile KJ, Handcock MS. Respondent-driven sampling: an assessment of current methodology. Sociol Methodol. 2010;40(1):285–327. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and/or analyzed during the current study may be available from the corresponding author on reasonable request.