Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Sep 1.
Published in final edited form as: J Subst Abuse Treat. 2019 Jul 13;104:144–147. doi: 10.1016/j.jsat.2019.07.006

Perceptions about Fentanyl-Adulterated Heroin and Overdose Risk Reduction Behaviors among Persons Seeking Treatment for Heroin Use

Michael D Stein 1, Shannon R Kenney 2, Bradley J Anderson 3, Genie L Bailey 4
PMCID: PMC6702957  NIHMSID: NIHMS1535464  PMID: 31370978

Abstract

Background:

Fentanyl-adulterated heroin supply chains are increasing risks for fatal overdose in the U.S.

Objective:

The current study examined the use of overdose risk reduction behaviors among persons seeking treatment for heroin use and whether perceptions about the presence of fentanyl in one’s heroin are associated with overdose risk reduction behaviors.

Method:

We recruited persons with opioid use disorder entering a managed withdrawal program. We used multiple linear regression to estimate the adjusted associations of participant characteristics and perception of fentanyl exposure with the frequency of engaging in each of five overdose reduction behaviors.

Results:

Participants (n=239; 75.3% male, 81.2% White, 67% injectors) estimated that 69.2% of the heroin they use contains fentanyl, and 94.6% knew that fentanyl increases overdose risk. Approximately 30% of respondents reported usually or always making sure others are around when they use heroin, carrying naloxone, taking “tester” doses of heroin or intentionally using in reduced amounts. While a majority of the sample reported never carrying naloxone or taking tester doses, and 70.2% reported never making sure that others around them carry naloxone, 84.5% had implemented one or more behavior at least rarely. Past month injection drug use was associated with making sure others are around, but perceptions about fentanyl in one’s heroin was not associated with use of harm reduction behaviors.

Conclusions:

In this sample of people who use heroin, although overdose risk reduction behaviors were not usually used, a majority had tried at least one behavior. That perceived exposure to fentanyl-adulterated heroin was not associated with the use of such behaviors provides important implications for public health education and intervention programming.

Keywords: Fentanyl, Heroin, Overdose, Harm Reduction

1.0. INTRODUCTION

Fentanyl, a synthetic opioid analgesic 30-50 times more powerful than heroin (Ciccarone, Ondocsin, & Mars, 2017) is the most commonly involved drug in overdose deaths in the U.S. (Hedegaard, Bastain, Trinidad, Spencer, & Warner, 2018). Fentanyl and fentanyl analogs have infiltrated national heroin supplies (Gladden, Martinez, & Seth, 2016; Warner, Trinidad, Bastian, Minino, & Hedegaard, 2016); of the estimated 72,000 drug overdose fatalities in 2017, the sharpest increase over the prior year occurred among 30,000 fentanyl-related deaths (National Institute on Drug Abuse, 2018).

This public health crisis is particularly concerning in Massachusetts. In the first six months of 2016, just prior to the current study’s data collection, 74% of available opioid overdose toxicology reports tested positive for fentanyl (Massachusetts Department of Public Health, 2017), and our own research in the state showed that 91% of patients entering short-term inpatient heroin withdrawal management tested positive for fentanyl (Kenney, Anderson, Conti, Bailey, & Stein, 2018).

Public health initiatives have focused on overdose risk reduction strategies in an effort to reduce burgeoning fentanyl-related overdose fatality rates. For example, naloxone kits, targeted at people who use opioids or potential bystanders and distributed through community-based programs and retail pharmacies offer a safe and effective antidote for reversing the effects of opioid overdose (Clark, Wilder, & Winstanley, 2014; Coffin & Sullivan, 2013; Doyon, Aks, & Schaeffer, 2014; Frank & Pollack, 2017; McDonald & Strang, 2016). Other strategies advocated in overdose risk reduction education include making sure others are around when using in case overdose occurs, reducing the amount of heroin used, and taking a partial or “tester” dose—a small “hit” or “taste”—prior to use as a way to perceive if fentanyl is present (e.g., New York City Municipal Drug Council, 2018; The Harm Reduction Coalition, 2018). Although not yet FDA-approved, rapid test strips, another risk reduction strategy that enables people to test for the presence of fentanyl in heroin, are becoming more widespread (Leins, 2018). However, research suggests that awareness about the prevalence of fentanyl and its overdose risk in personal supply chains may not alter heroin-using behaviors (Ciccarone et al., 2017; Somerville et al., 2017).

1.1. Study Goals and Hypotheses

Public harm reduction policy and clinical experience suggests that persons who use heroin may implement at least five strategies to reduce their overdose risk. These include using a “tester” dose of heroin before a use episode, although the accuracy of such “testing” remains unclear; using smaller amounts of heroin; using only in the presence of others; carrying naloxone (e.g., naloxone administration; Kenney, Anderson, Bailey, & Stein, 2018); and checking that others nearby are carrying naloxone, hence ensuring its availability in the case of an overdose. This study is the first to assess, in a single cohort, the use of these five overdose reduction behaviors by people using heroin. The current study aims to assess the frequency of overdose risk reduction behaviors and shed light on the promise of public health education that traditionally aims to raise awareness in order to promote behavior change. We hypothesized that an individual’s perception about the presence of fentanyl-adulterated heroin in their supplies would be associated with risk reduction behaviors. When considering these five behaviors, we controlled for history of medication-assisted treatment (MAT), past month injection drug use (IDU), and past year overdose, which are associated with fentanyl exposure and risk reduction behaviors (Arfken, Suchanek, & Greenwald, 2017; Hayashi et al., 2018; Kenney, Anderson, Conti, et al., 2018).

2. Material and methods

2.1. Recruitment

The current sample was drawn from patients seeking inpatient opioid withdrawal management at Stanley Street Treatment and Resources, Inc. (SSTAR) in Fall River, Massachusetts. SSTAR provides evaluation, supervised symptom management using a methadone taper schedule, individual and group education and counseling, and aftercare case management (mean length-of-stay is 5.2 days).

A total of 318 patients were admitted to SSTAR during the recruitment period (October 2017-May 2018). Patients 18 years or older, English-speaking, and able to provide informed consent as approved by the Butler Hospital Institutional Review Board were invited to participate in a survey research study. Twenty-six refused study participation or were discharged before staff could perform an interview. Of 292 persons completing a 15-minute non-incentivized, face-to-face interview administered by non-treating research staff, 27 were missing responses to the harm reduction items. The 239 (81.9%) persons who reported last month heroin use constituted the study sample.

2.2. Measures

In addition to age, gender, years of education, and IDU in the past month we assessed the following variables.

2.2.1. Past year overdose.

Respondents were asked if they had overdosed in the past year. We defined overdose (on any drug) as “you were unarousable (couldn’t be woken) with shaking or calling your name because of the drugs you consumed.”

2.2.2. History of Medication-Assisted Treatment.

Respondents were asked if they had ever been prescribed buprenorphine or naltrexone or been enrolled in a methadone program. Persons answering “yes” to any of these were considered to a history of MAT.

2.2.3. Fentanyl exposure perceptions.

Respondents were asked if they think fentanyl increases risk for overdose: “yes,” “no,” or “don’t know.”

2.2.4. Perceptions about fentanyl-adulterated heroin.

Respondents were asked to estimate what percent (from 0-100%) of heroin they use has fentanyl in it.

2.2.5. Overdose risk reduction behaviors.

Respondents were asked how often they engage in five behaviors while using heroin, on a six-point ordered-response scale: “never” to “always.” Behaviors included “making sure other people are around,” “use in small doses,” “carrying naloxone (narcan),” “making sure others around them have naloxone (narcan) with them,” “taking a small hit first (a tester shot).”

2.2.6. Knowledge about fentanyl test strips.

Respondents were asked if they had “ever heard of fentanyl test strips (thin piece of plastic that can be dipped in a drug sample) that you can use to see, within five minutes, if fentanyl is in a batch of heroin before you use.” Response options were “yes” or “no.” Respondents familiar with fentanyl test strips were asked how often they engaged in “testing drugs before using with a rapid test strip.”

2.3. Analytical Methods

We present descriptive statistics to summarize sample characteristics and frequency of engaging in harm reduction behaviors. We used multiple linear regression to test the hypothesis that perceived percentage of heroin containing heroin is positively associated with increased use of harm reduction behaviors after adjusting for background characteristics, history of MAT, past month IDU, and past year overdose. We report fully-standardized regression coefficients for continuous covariates, and y-standardized coefficients for categorical covariates. All tests of significance were based on the Huber-White variance estimator which is robust to estimate heteroskedasticity.

We conducted auxiliary analysis to determine if past month IDU moderated the association between perceived percentage of heroin containing fentanyl and harm reduction behaviors. These were tested by including the first-order interaction of these two variables in a linear regression model.

3. Results

Participants averaged 33.6 (± 8.8) years of age, 75.3% were male, 81.2% were White, and 11.3% were Latino/a. Mean years of education was 11.8 (± 1.70). About 76.0% had a lifetime history of MAT, 65.9% reported IDU in the past 30 days, and 37.2% reported a drug overdose within the past year. On average, participants estimated that 69.2% (± 32.1, Median = 80%, Range 0% - 100%) of the heroin they use contains fentanyl. In all, 94.6% of respondents believed fentanyl increases overdose risk; this variable was omitted from subsequent analyses because of little variation.

Distributions of individual harm reduction items are given in Table 1. On an ordered response scale of 0-5, mean and median scores on the harm reduction behaviors were at or near the lower limit. Nearly half (47.7%) said they never made sure others were around when they used. Moreover, 51.9% said they never carried naloxone and 70.2% said they never made sure other persons around them were carrying naloxone. Over half (51.3%) said they never took tester shots and 43.1% said they never used in small doses. Thirty-seven participants (15.5%) answered “never” to all five overdose reduction behaviors; that is, 84.5% had tried one or more behavior at least “rarely.” Twenty-eight participants (11.7%) reported they had heard of fentanyl test strips and only three of these respondents said they ever used rapid test strips. Because of the very low frequency of endorsement, this behavior was not evaluated in subsequent analyses.

Table 1.

Response Distributions on Harm Reduction Behavior Items (n = 239).

Never Rarely Occasionally Sometimes Usually Always
Item Mean (± SD) Median 0 1 2 3 4 5
I make sure other people are around when I use. 1.8 (± 2.0) 1 47.7% 10.5% 5.0% 7.1% 13.0% 16.7%
I carry naloxone (narcan) with me. 1.8 (± 2.1) 0 51.9% 8.0% 1.7% 7.5% 10.5% 20.5%
I make sure others who are around me have naloxone (narcan) with item. 0.9 (± 1.6) 0 70.2% 7.6% 4.2% 5.5% 5.9% 6.7%
I take a small hit first (a tester shot). 1.8 (± 2.1) 0 51.3% 5.5% 3.4% 8.0% 13.9% 18.1%
I use in small doses. 1.9 (± 2.0) 1 43.1% 10.0% 4.6% 12.1% 15.9% 14.2%
Open in a new tab

The data did not support our hypothesis that the perceived percentage of heroin containing fentanyl would be associated with increased use of harm reduction behaviors (Table 2). After adjusting for background characteristics, history of MAT, past month IDU, and past year overdose, the associations of perceived percentage of heroin containing fentanyl with harm reduction behaviors were not statistically significant, and not directionally consistent across all harm reduction behaviors.

Table 2.

Multiple Linear Regression Models Estimating the Adjusted Associations of Background Characteristics, Past Month IDU, Past Year Overdose, and Perception of the Proportion of Heroin that Contains Fentanyl.

HARM REDUCTION OUTCOME
Make Sure
Others Are
Around		 Carry Naloxone Others Around
Carrying
Naloxone		 Take Tester
Shot		 Use in Small
Doses
βyx (SE)a βyx (SE)a βyx (SE)a βyx (SE)a βyx (SE)a
Years Age −.013 (0.072) −0.101 (0.062) −.0.077 (0.072) −.0.117 (0.073) 0.025 (0.075)
Sex (Male) −.0.235 (0.172) −.0.526** (0.166) −.0.005 (0.192) 0.151 (0.170) 0.014 (0.156)
Years Education −.0.022 (0.076) 0.131 (0.075) 0.137 (0.079) 0.189** (0.073) 0.094 (0.076)
Ever MAT −.0.301 (0.081) 0.253 (0.147) 0.186 (0.138) 0.182 (0.151) −.0.233 (0.165)
Past Month IDU 0.344* (0.146) 0.205 (0.137) 0.130 (0.141) −.0.091 (0.146) 0.013 (0.149)
Past Year Overdose 0.172 (0.148) 0.268 (0.138) 0.229 (0.165) 0.028 (0.141) −.0.213 (0.139)
Perceived %w Fentanyl −.0.010 (0.067) 0.060 (0.058) 0.050 (0.071) −.0.062 (0.068) −.0.030 (0.070)
Constant 0.008 −.0.034 −.0.307* −.0.187 0.239
Observations 239 239 238 238 239
Open in a new tab
*

p < .05,

**

p < .01,

***

p < .001

a

Coefficients are fully standardized for continuous covariates and y-standardized for categorical covariates. Robust Huber-White standard errors.

A few of the associations evaluated in this study were statistically significant (Table 2). Males were significantly less likely than females to say they carried naloxone with them (βyx = −0.53, p < .01). Education was positively associated with frequency of taking tester shots (βyx = 0.19, p < .05). Past month IDU was positively and significantly associated with making sure others were around when using drugs (βyx = 0.34, p < .05).

We also estimated models to determine if past month IDU moderated the association between perceived percentage of heroin containing fentanyl and harm reduction behaviors. The mean perceived percentage of heroin containing fentanyl was significantly (t236 = −2.57, p = .011) higher among those who had recently injected drugs (72.7, ± 30.0) than non-injectors (61.5, ± 34.8). However, there was no statistically significant evidence that the relationship between perceptions about fentanyl and harm reduction behaviors was moderated by recent IDU; p-values for all tests of first-order interaction were > .20.

4. Discussion

Consistent with extant research (Macmadu, Carroll, Hadland, Green, & Marshall, 2017), this sample showed a strong awareness that fentanyl exposure increases overdose risk. Yet even in the midst of an overdose epidemic, overall, respondents did not report implementing overdose risk reduction behaviors “usually.” A majority of respondents reported never taking tester shots or carrying naloxone, and a majority (70.2%) report never ensuring others carry naloxone when they use heroin. On the other hand, a majority of participants (84.5%) reported using one or more harm reduction behaviors at least “rarely,” with approximately one in three respondents reporting usually or always making sure others are around, personally carrying naloxone, taking tester hits, and using in small doses when using heroin.

The current study does not support our hypothesis that respondent’s perceptions about the presence of fentanyl in their heroin impacts their use of overdose risk reduction behaviors. Rather, these results support prior research showing that increasing awareness about overdose risks may have limited impact on promoting protective drug use behaviors among people who use heroin (Kerr, Small, Hyshka, Maher, & Shannon, 2013). In fact, even though respondents reporting (versus not reporting) past month IDU were more likely to report making sure others are around when they use and perceive higher levels of fentanyl in their heroin, IDU did not moderate the relationship between these perceptions and the risk behaviors assessed in this study. These findings indicate that leveraging general education about fentanyl risks alone may be ineffective. More intensive interventions that educate people about harm reduction behaviors while targeting underlying cognitive or psychological barriers associated with substance misuse (cognitive behavioral therapy; Rohsenow, 2016) or that establish goals for change and enhance motivation to modify behaviors (motivational interviewing; R. Miller & Rollnick, 2002) may be warranted. Using peers who are pursuing self-protective behaviors to train other injectors could perhaps also change community standards. Still, to date, structural and environmental programs, such as MAT (Connery, 2015 for review) and naloxone distribution (Clark et al., 2014; McDonald & Strang, 2016), are demonstrated to be most effective in reducing overdose risk.

It is important to note that given the unique effects of fentanyl, which rapidly accelerate the risk of overdose death, it is unclear how effective traditional strategies may be in preventing overdose. For example, the most commonly endorsed risk reduction behavior assessed was using heroin in small doses (30.1% reported engaging in this behavior usually or always). However, a miniscule dose of fentanyl can lead to overdose, putting into question the effectiveness of small doses or even tester hits. Moreover, standard naloxone dosing is less effective in cases of fentanyl overdose. People perceiving that fentanyl is involved in an overdose may also believe that certain harm reduction strategies are ineffective and, in turn, not use such strategies. Research that accounts for this complex interplay between fentanyl-related perceptions and beliefs is needed.

Only a few people in this sample used rapid test strips. This finding likely reflects the novelty of rapid test strips that are not yet available through syringe exchange programs or for consumers to purchase in Massachusetts. In fact, only 28 (11.7%) respondents had ever heard of fentanyl test strips. Examining the use of rapid test strips in other states and as they become more widely disseminated may be an important avenue for future investigation.

4.1. Limitations

This study has several limitations. First, respondents were seeking heroin withdrawal treatment at a single site and might not apply to individual not in treatment, those receiving maintenance opioid use disorder medication, or prescription opioid misusers. Second, data were based on self-reports which may not reflect actual harm reduction behaviors (P. G. Miller, 2007). Moreover, although there was no specific time frame offered for the harm reduction questions, questions were posed in the present tense for these active heroin users. Therefore, we presume that responses capture current practices and not lifetime use of harm reduction behaviors, for example. Third, an overarching assumption in this study was that respondents knew about these risk reduction strategies. It is possible that some participants were not aware of these potential overdose reduction strategies. Conversely, it may be the case that some participants were aware that two strategies (tester shots and small doses) may not necessarily reduce overdose risk and therefore they chose not to use these strategies. Fourth, respondents reported their use of protective behaviors but not their desire to avoid fentanyl. Research demonstrates that some people who use heroin try to avoid fentanyl (Carroll, Marshall, Rich, & Green, 2017), while others are ambivalent or intentionally use fentanyl-adulterated heroin (Ciccarone et al., 2017; Macmadu et al., 2017). Fifth, other covariates that might moderate the outcomes such as duration of heroin use, fatalism, impulsivity, and education about harm reduction behaviors, were not measured. Finally, it is beyond the scope of these data to assess the effectiveness of these strategies in preventing overdose.

4.2. Conclusion

In this sample of persons using heroin, a majority believed their heroin contained fentanyl and while a majority had at least tried one of the assessed harm reduction behaviors, only a minority always used protective behaviors. Our findings suggest that “universal precautions” against overdose by people who use heroin is a distant hope. Reminding persons who continue to use heroin about these overdose risk reduction behaviors and enhancing their self-efficacy to implement protective behaviors could reduce overdose mortality. But until heroin users’ norms about using these protective behaviors change, this population will remain at high risk of overdose death. To date, entering and maintaining effective MAT for opioid use disorder remains the preferred course of care for this at-risk population.

Highlights.

  • Treatment-seeking heroin using participants estimated that most heroin they use (69.2%) contains fentanyl.

  • Nearly all (94.6%) participants knew that fentanyl increases overdose risk.

  • 15.5% of participants reported never using any of the assessed overdose reduction behaviors.

  • A majority (84.5%) had used one or more behavior at least “rarely.”

  • Beliefs about the presence of fentanyl in personal heroin supplies was not associated with frequency of individual overdose reduction behaviors.

Acknowledgments

This study was funded by the National Institute on Drug Abuse (RO1 DA034261). Trial registered at clinicaltrials.gov; Clinical Trial # .

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Arfken CL, Suchanek J, & Greenwald MK (2017). Characterizing fentanyl use in methadone-maintained clients. J Subst Abuse Treat, 75, 17–21. doi: 10.1016/j.jsat.2017.01.004 [DOI] [PubMed] [Google Scholar]
  2. Carroll JJ, Marshall BDL, Rich JD, & Green TC (2017). Exposure to fentanyl-contaminated heroin and overdose risk among illicit opioid users in Rhode Island: A mixed methods study. Int J Drug Policy, 46, 136–145. doi: 10.1016/j.drugpo.2017.05.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ciccarone D, Ondocsin J, & Mars SG (2017). Heroin uncertainties: Exploring users' perceptions of fentanyl-adulterated and −-substituted 'heroin'. Int J Drug Policy, 46, 146–155. doi: 10.1016/j.drugpo.2017.06.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Clark AK, Wilder CM, & Winstanley EL (2014). A systematic review of community opioid overdose prevention and naloxone distribution programs. J Addict Med, 8(3), 153–163. doi: 10.1097/ADM.0000000000000034 [DOI] [PubMed] [Google Scholar]
  5. Coffin PO, & Sullivan SD (2013). Cost-effectiveness of distributing naloxone to heroin users for lay overdose reversal. Ann Intern Med, 158(1), 1–9. doi: 10.7326/0003-4819-158-1-201301010-00003 [DOI] [PubMed] [Google Scholar]
  6. Connery HS (2015). Medication-assisted treatment of opioid use disorder: review of the evidence and future directions. Harv Rev Psychiatry, 23(2), 63–75. doi: 10.1097/HRP.0000000000000075 [DOI] [PubMed] [Google Scholar]
  7. Doyon S, Aks SE, & Schaeffer S (2014). Expanding access to naloxone in the United States. J Med Toxicol, 10(4), 431–434. doi: 10.1007/s13181-014-0432-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Frank RG, & Pollack HA (2017). Addressing the Fentanyl Threat to Public Health. N Engl J Med, 376(7), 605–607. doi: 10.1056/NEJMp1615145 [DOI] [PubMed] [Google Scholar]
  9. Gladden RM, Martinez P, & Seth P (2016). Fentanyl law enforcement submissions and increases in synthetic opioid-involved overdose death-27 sates, 2013–2014. Morbidity and Mortality Weekly Report, 65(33), 837–843. [DOI] [PubMed] [Google Scholar]
  10. Hayashi K, Milloy MJ, Lysyshynd M, DeBecka K, Nosovaa E, Wood E, & Kerr T (2018). Substance use patterns associated with recent exposure to fentanyl among people who inject drugs in Vancouver, Canada: A cross-sectional urine toxicology screening study. Drug And Alcohol Dependence, 183, 1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Hedegaard H, Bastain BA, Trinidad JP, Spencer M, & Warner M (2018). Drugs most frequently involved in drug overdose deaths: United States, 2011–2016. National Vital Statistics Reports, 67(9), 1–13. [PubMed] [Google Scholar]
  12. Kenney SR, Anderson BJ, Bailey GL, & Stein MD (2018). Factors associated with naloxone administration in an opioid dependent sample. J Subst Abuse Treat, 84, 17–20. doi: 10.1016/j.jsat.2017.10.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Kenney SR, Anderson BJ, Conti MT, Bailey GL, & Stein MD (2018). Expected and actual fentanyl exposure among persons seeking opioid withdrawal management. J Subst Abuse Treat, 86, 65–69. doi: 10.1016/j.jsat.2018.01.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kerr T, Small W, Hyshka E, Maher L, & Shannon K (2013). 'It's more about the heroin': injection drug users' response to an overdose warning campaign in a Canadian setting. Addiction, 108(7), 1270–1276. doi: 10.1111/add.12151 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Leins C (2018, June 1). California funds fentanyl test strips. U.S. News & World Report. [Google Scholar]
  16. Macmadu A, Carroll JJ, Hadland SE, Green TC, & Marshall BD (2017). Prevalence and correlates of fentanyl-contaminated heroin exposure among young adults who use prescription opioids non-medically. Addict Behav, 68, 35–38. doi: 10.1016/j.addbeh.2017.01.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Massachusetts Department of Public Health. (2017). Data Brief: Opioid-related Overdose Deaths Among Massachusetts Residents. Retrieved from [Google Scholar]
  18. McDonald R, & Strang J (2016). Are take-home naloxone programmes effective? Systematic review utilizing application of the Bradford Hill criteria. Addiction, 111(7), 1177–1187. doi: 10.1111/add.13326 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Miller PG (2007). Media reports of heroin overdose spates: Public health messages, moral panics or risk advertisements? Critical Public Health, 17, 113–121. [Google Scholar]
  20. Miller R, & Rollnick S (2002). Motivational interviewing: Preparing people for change (2nd ed.). New York: Guilford Press. [Google Scholar]
  21. National Institute on Drug Abuse. (2018). Overdose Death Rates. . National Institutes of Health; Retrieved from https://www.drugabuse.gov/related-topics/trends-statistics/overdose-death-rates. [Google Scholar]
  22. New York City Municipal Drug Council. (2018). New York City Municipal Drug Strategy Council: 2018 Report and Recommendations. Retrieved from New York, NY: [Google Scholar]
  23. Rohsenow DJ (2016). Cognitive behavioral therapy for substance abuse/dependence In Litz H. F. a. B. (Ed.), Encyclopedia of Mental Health (2nd ed., pp. 307–316). [Google Scholar]
  24. Somerville NJ, O'Donnell J, Gladden RM, Zibbell JE, Green TC, Younkin M, … Walley AY (2017). Characteristics of Fentanyl Overdose - Massachusetts, 2014–2016. MMWR Morb Mortal Wkly Rep, 66(14), 382–386. doi: 10.15585/mmwr.mm6614a2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. The Harm Reduction Coalition. (2018). Opioid Overdose basics. Retrieved from https://harmreduction.org/issues/overdose-prevention/overview/overdose-basics/
  26. Warner M, Trinidad JP, Bastian BA, Minino AM, & Hedegaard H (2016). Drugs Most Frequently Involved in Drug Overdose Deaths: United States, 2010-2014. Natl Vital Stat Rep, 65(10), 1–15. [PubMed] [Google Scholar]

RESOURCES