Abstract
Background:
Nonpharmaceutical fentanyl (NPF) is driving the national epidemic of opioid overdose deaths. Clinicians can play a role in fostering awareness of this growing risk and delivering interventions to reduce mortality. However, there is limited research assessing clinician knowledge, attitudes, and practices relating to NPF and harm reduction strategies.
Methods:
A 34-question survey was designed to assess knowledge, attitudes, and practices related to NPF and harm reduction strategies of adult and pediatric hospital-based and emergency clinicians at a single academic medical center. Results were summarized using descriptive statistics. Chi square and Fishers exact tests were used to compare groups.
Results:
There were 136 survey responses. The majority (88%) of respondents correctly answered a question on NPF potency. Most respondents were aware that NPF exposure was very (84%) or somewhat likely (10%) for someone using illicit opioids and very (44%) or somewhat likely (46%) for nonopioid drugs. Respondents viewed overdose prevention as highly important for patients using illicit opioids (93%) and nonopioid drugs (86%) but few (21%) were very/extremely familiar with overdose prevention strategies and just over half (57%) were comfortable/very comfortable counseling about overdose prevention. There was wide variability in utilization of harm reduction/treatment strategies (7.3% frequently providing fentanyl test kits to 70% frequently prescribing naloxone). Higher levels of comfort and familiarity with overdose prevention were associated with more frequent counseling on harm reduction strategies. Pediatric-only clinicians had less familiarity (5% very/extremely familiar) and comfort (35% comfortable/very comfortable) with overdose prevention, and limited use of harm reduction strategies (0%-31% using each strategy frequently).
Conclusions:
While clinicians had knowledge and awareness of NPF and rated overdose prevention as highly important, utilization of harm reduction and treatment strategies was variable. This study highlights opportunities for education and system-based support to improve clinician-driven harm reduction practices for patients at risk of overdose.
Highlights
While clinicians rated overdose prevention as highly important, use of harm reduction strategies to reduce overdose risk when caring for at-risk patients was variable and often limited.
Clinicians’ limited levels of comfort and familiarity with overdose prevention points to opportunities for interventions to increase utilization of harm reduction and treatment strategies.
Given the rising rates of adolescent overdoses, it is particularly important to highlight the gap in pediatric clinician familiarity, comfort, and utilization of overdose prevention strategies. Future initiatives should target this population for intervention.
Introduction
Nonpharmaceutical fentanyl (NPF), an illicitly manufactured high-potency synthetic opioid, is currently the main driver of the national epidemic of opioid overdose deaths. In 2021, the age-adjusted rate of overdose deaths involving NPF was twice as high as any other opioid or stimulant drug. 1 NPF has largely replaced heroin in the illicit drug market and can be found in other drugs, including stimulants and counterfeit prescription medications.2,3 Accidental ingestion of NPF has led to increasing overdoses among pediatric and adolescent populations, and adolescent overdose mortality has increased more than 200% between 2019 and 2021.4,5 Understanding the unique risks associated with NPF and being able to effectively communicate harm reduction strategies, which encompasses several measures to prevent overdose including carrying naloxone, using fentanyl test kits, reducing illicit drug use, and not using alone, is key to decreasing overdose deaths.
Clinicians can play a role in fostering awareness of harm reduction strategies to prevent opioid overdoses. The Substance Abuse and Mental Health Services Administration advises clinicians to provide overdose prevention education, prescribe naloxone to patients at risk of opioid-related harms, and offer treatment to patients with opioid use disorder (OUD). 6 However, these interventions are often not provided to people who use drugs (PWUD) seeking healthcare.7,8 People at high risk of overdose are often not prescribed naloxone despite contact with healthcare systems, and clinician naloxone prescribing is variable even for patients identified to be at high risk of overdose.9-12
Despite the increased risk that NPF poses, clinician knowledge and practices related to NPF are not well described. To leverage provider capacity to reduce overdose deaths, it is important to uncover clinician-level factors contributing to low uptake of overdose prevention interventions. It is also vital to identify how clinicians have adapted their practices when treating PWUD in the setting of rising overdose deaths and NPF prevalence. Through a survey of adult and pediatric clinicians in various practice settings, this study aimed to describe clinician knowledge of and attitudes toward NPF and overdose prevention as well as current approaches to managing patients at risk of overdose.
Methods
Survey Development
A 34-question survey was designed to assess clinician knowledge, attitudes, and practices related to NPF and harm reduction strategies that can reduce the risk of death from opioid overdose. The survey development was informed by the Precede-Proceed implementation science framework and a review of the literature on the characteristics of NPF, experiences of PWUD with fentanyl, and clinician knowledge and attitudes toward OUD treatment, naloxone prescribing, and overdose prevention.11,13-22 Survey questions were iteratively refined through input from local clinicians with subject matter expertise and pilot testing.
Respondents and Procedures
The survey was conducted at a large, urban academic medical center in the Midwest from May to June 2023. Surveys were distributed to attending physicians and advanced practice providers from the sections of hospital medicine, emergency medicine, pediatric hospital medicine, and pediatric emergency medicine. Resident physicians from the internal medicine, emergency medicine, medicine-pediatrics, and pediatric residency programs were also eligible to complete the survey. Anonymous surveys were distributed by email, at educational conferences, and in workrooms. Data were collected and managed using REDCap.23,24 The study was determined to be exempt by the Institutional Review Board.
Measures
The survey instrument (Supplemental Material 1) evaluated provider knowledge of NPF and the impact of NPF on clinical practice. The survey assessed attitudes toward overdose prevention, asking respondents to rate the importance of overdose prevention, familiarity with overdose prevention strategies, and comfort discussing overdose prevention on 5-point Likert scales. To assess clinician practices, respondents reported their use of harm reduction (naloxone, fentanyl test kits, safer drug use counseling) and treatment (initiation of buprenorphine, initiation of methadone, referral to substance use disorder (SUD) treatment) strategies when caring for patients they determine to be at risk of overdose on a 5-point Likert scale of frequency. Demographic information and practice characteristics were collected and included questions on frequency of clinical exposure to opioid overdose and patients with OUD.
Statistical Analysis
Likert scale responses were dichotomized for the following measures: importance of overdose prevention (low importance = not at all/slightly/somewhat important, high importance = very/extremely important), familiarity with overdose prevention strategies (low familiarity = not at all/slightly/somewhat familiar, high familiarity = very/extremely familiar), comfort discussing overdose prevention (low comfort = very uncomfortable/uncomfortable/neutral, high comfort = comfortable/very comfortable), harm reduction/treatment strategy use (infrequent = never/rarely/sometimes, frequent = often/always; responses of “not relevant to my clinical practice” were excluded from analysis). The sample was first characterized using descriptive statistics. Chi square test or Fishers exact test were used to test differences between groups. Analyses were performed using R version 4.0 (R Core Team, 2020). 25
Results
Of 417 eligible respondents, 136 (33%) completed the survey. Residents made up 53% of responses (N = 72, response rate = 30%), attending physicians 32% (N = 43, response rate = 32%), and advanced practice providers 14% (N = 19, response rate = 48%). Fifteen percent of respondents (N = 20, response rate = 20%) cared for only pediatric patients. Twenty-one percent (N = 29) primarily practiced in the emergency department. Sixty-five percent (N = 89) cared for patients with OUD more than once per week (high clinical exposure to OUD). Twenty-four percent (N = 32) cared for patients more than once per week who presented after an opioid overdose (high clinical exposure to overdose; Table 1).
Survey Respondent Demographics (N = 136).
Abbreviation: OUD, opioid use disorder.
Almost all respondents (88%) knew that NPF is more potent than heroin. Sixty percent responded that overdoses involving NPF may require more naloxone and 40% responded that NPF use may make buprenorphine inductions more difficult (46% answered “don’t know”). Respondents perceived there to be a high risk of exposure to NPF for someone using illicit opioids with 84% rating exposure as very likely. There was a lower perceived risk of exposure to NPF for someone using illicit nonopioid drugs, 44% rating it very likely and 46% somewhat likely.
Respondents largely rated overdose prevention as highly important for patients using illicit opioids and nonopioid drugs, 93% and 86%, respectively. Just over half of the respondents (57%) reported being comfortable discussing preventing overdose with patients. Fewer respondents (21%) reported high familiarity with overdose prevention strategies (Table 2).
Attitudes Toward Overdose Prevention (N = 136).
The most frequently used harm reduction or treatment strategy when caring for patients at risk of overdose was naloxone prescribing. For patients at risk of overdose, 70% of clinicians frequently (often/always) prescribed naloxone, 59% frequently counseled about naloxone use, and 30% frequently recommended over-the-counter naloxone. Twenty-nine percent of respondents frequently counseled about strategies for safer drug use and 12% frequently recommended using fentanyl test kits (Figure 1). Twenty-three percent frequently initiated buprenorphine, and 18% of inpatient clinicians frequently initiated methadone. Forty-nine percent of respondents answered that a patient testing positive for fentanyl would make them more likely to utilize harm reduction strategies and 34% responded that it would not change the likelihood. Of those who responded that it would not change the likelihood, 90% frequently prescribed naloxone and 71% frequently counseled on naloxone. The most common barriers to utilizing harm reduction strategies were lack of time (70%), lack of knowledge (56%), lack of comfort (35%), and patient acceptability (41%).

Clinician use of harm reduction and treatment strategies for patients at risk of overdose (N = 136).
Increased comfort and familiarity with overdose prevention were associated with more frequent implementation of harm reduction and treatment strategies. Clinicians with high levels of comfort discussing overdose prevention used several strategies more frequently than those with lower comfort levels: counseling on strategies for safer use (44% vs 6%, P < .001), counseling about naloxone use (73% vs 40%, P < .001), recommending over-the-counter naloxone (37% vs 20%, P = .04), and referring patients to substance use treatment (60% vs 37%, P = .009; Table 3). Higher reported familiarity with overdose prevention strategies was associated with more frequent counseling on strategies for safer drug use (54% vs 21%, P < .001) and counseling about naloxone use (79% vs 54%, P = .02; Table 3). High clinical exposure to OUD was associated with increased naloxone prescribing (77% vs 54%, P = .01) but no other interventions. There was no association between clinical exposure to opioid overdose and utilization of harm reduction or OUD treatment interventions.
Association Between Clinician Comfort/Familiarity With Overdose Prevention and Frequent Use of Harm Reduction and Treatment Strategies (N = 136).
Counts and percentages represent the respondents in each group who reported using each strategy frequently (often/always) when caring for patients at risk of overdose, excluding those who responded that the strategy was not relevant to their clinical practice.
Among respondents who care only for pediatric patients, familiarity and comfort with overdose prevention strategies were limited. Only 5% reported having high levels of familiarity with overdose prevention strategies and 35% had high levels of comfort counseling on overdose prevention. Thirty-one percent of pediatric clinicians counseled on naloxone use and 8% prescribed naloxone frequently for patients at risk of overdose. An additional 35% to 45% of pediatric clinicians reported that each strategy was not relevant to their clinical practice. The most frequently reported barrier among pediatric clinicians was lack of knowledge (60%).
Discussion
Emergency and hospital-based clinicians had general knowledge and awareness of NPF and viewed overdose prevention strategies as important, but their utilization of harm reduction and treatment strategies was variable. Among pediatric clinicians, comfort, familiarity, and utilization of harm reduction strategies were low.
The only harm reduction or treatment strategy utilized consistently by clinicians in this study was prescribing and counseling on naloxone. Rates of reported naloxone prescribing were higher than seen in published literature,11,21 possibly because of temporal trends of increasing awareness among clinicians, as well as local interventions to increase naloxone prescribing including an electronic health record alert. There is widespread consensus that naloxone distribution to patients at risk of experiencing or witnessing an overdose is critical to reduce opioid-related mortality. Clinicians who treat these patients should be distributing naloxone routinely. System-based interventions such as improving ease of naloxone prescribing or offloading screening and counseling to nonclinician staff,26,27 could be impactful given that 30% of respondents prescribed it infrequently. Clinicians infrequently provided counseling on or access to fentanyl test kits, highlighting an opportunity for this intervention to detect fentanyl presence in illicit nonopioid drugs that increasingly contain NPF. Strategies for safer drug use (eg, not using drugs alone, using less, using more slowly) are infrequently used by PWUD14,28 and there is an opportunity for clinicians to raise awareness about these harm reduction interventions. However, clinician counseling on these strategies was infrequent.
Consistent with the reported experiences of PWUD,14,29-31 the vast majority of respondents perceived there to be a high likelihood of NPF exposure for someone using illicit opioids. Despite this and the belief that overdose prevention is highly important in the context of high regional rates of overdose, 1 implementation of harm reduction strategies beyond naloxone prescribing was limited and inconsistent. Limited clinician knowledge, for example, on topics like the impact of NPF use on buprenorphine induction, may contribute to their infrequent implementation of interventions. This study identifies a need for broader and more advanced education and systems to support the many ways clinicians can impact patients at risk for overdose. To close this gap, targeted education should aim to improve familiarity and comfort among clinicians who recognize the importance of harm reduction but practice evidence-based strategies less frequently. Clinicians also frequently reported lack of time as a barrier. This highlights the need to improve knowledge of harm reduction among all staff so clinicians are able to more effectively provide education in limited time and other members of the clinical team are also equipped to offer education on overdose prevention strategies.
Interventions are especially needed for pediatric clinicians, as they care for adolescents, a group experiencing a rapid increase in overdose mortality. Since 2020 adolescent overdose mortality has increased at a rate greater than that of the overall population.4,5 Despite believing that overdose prevention was very important, few pediatric clinicians in this study frequently prescribed naloxone or counseled on strategies to reduce overdose risk. Many reported that the strategies were not relevant to their clinical practice, perhaps reflecting a lack of awareness. Pediatric clinicians had low levels of familiarity and comfort with overdose prevention and frequently reported lack of knowledge as a barrier, consistent with previously reported barriers to pediatric resident naloxone prescribing. 32 Limited education on overdose prevention in pediatric training may also contribute to these findings. 33 In the context of rising adolescent overdoses, it is important to address these barriers for pediatric clinicians.
Limitations of this study include survey distribution at a single institution and response rate. Clinicians who found the topic of the survey less relevant may have chosen not to respond, creating a selection bias by eliminating respondents with less exposure to the management of OUD and opioid overdoses. This is particularly relevant for pediatric clinicians, who responded at a lower rate than clinicians who care for adult patients. Most respondents cared for patients with OUD relatively frequently, which may limit the generalizability of the data. The cross-sectional nature does not allow identification of causal relationships, and self-report measures may not reflect the actual practices of clinicians. Measures were dichotomized from Likert scales, which may cause a loss of variability. Finally, while the survey was pilot tested and iteratively refined, survey measures were not formally tested for validity or reliability.
Conclusions
This study described ED- and hospital-based clinician knowledge, attitudes, and practices relating to NPF and harm reduction strategies. We identified a gap between the clinician perceived importance of overdose prevention and the implementation of overdose reduction strategies. These results highlight opportunities for education and system-based support to improve clinician comfort and familiarity needed to practice harm reduction and reduce patient risk due to NPF.
Supplemental Material
sj-docx-1-saj-10.1177_29767342241266421 – Supplemental material for Understanding Clinician Knowledge, Attitudes, and Practices Relating to Nonpharmaceutical Fentanyl and Harm Reduction
Supplemental material, sj-docx-1-saj-10.1177_29767342241266421 for Understanding Clinician Knowledge, Attitudes, and Practices Relating to Nonpharmaceutical Fentanyl and Harm Reduction by Elena Whitney, George Weyer, Molly Perri, Sarah Dickson, Angela Kerins, Andrea Justine Landi, P. Quincy Moore, John P. Murray, Geoff Pucci and Mim Ari in Substance Abuse
Footnotes
Author Contributions
All authors contributed to the study design. EW and MA performed the data collection and analysis. EW, MA, and GW contributed in the data interpretation and writing of the manuscript. All authors provided critical revisions to the manuscript. All authors provided final approval of the version to be published.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the University of Chicago Pritzker School of Medicine.
Compliance,Ethical Standards,and Ethical Approval
The study was determined to be exempt by the University of Chicago Institutional Review Board.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
