Research Analysis: Conclusions about 'moral hazard' of naloxone not supported by methodology
Further research is needed to determine if naloxone access laws actually increase distribution and use of naloxone, and high-risk behavior by patients with opioid use disorder
By Ryan Marino, Brian Fullgraf and Jeremiah Escajeda
Economics researchers recently released, "The Moral Hazard of Lifesaving Innovations: Naloxone Access, Opioid Abuse, and Crime," a working paper on an open access journal website. The paper has received much attention from mass media outlets and EMS professional publications.
The research paper's authors state that increased naloxone distribution resulted in increased crime and opioid abuse with no change in mortality, as well as concluding that increased naloxone distribution resulted in increased fentanyl use. The working paper bypassed the traditional peer-review and the authors chose to publicize their findings on Twitter.
The authors used econometric models to evaluate changes in opioid-related emergency department admissions, mortality and thefts for one year after a state passed naloxone access legislation. The study then compared data sets from states where legislation was passed with those states not yet passing naloxone access legislation, and compared trends over time.
The authors conclude that during the study period, in relation to naloxone access laws, there was an increase in arrest rates for opioids, broadened sales of fentanyl and that naloxone access laws resulted in users seeking "higher highs" by using more fentanyl. The study concludes that these naloxone access laws have no significant impact on opioid-related mortality, and specifically worsened opioid mortality in the midwestern and southern regions of the United States.
Memorable quotes on naloxone access, opioid abuse and crime
Here are three memorable quotes from the research paper, which can be read in full at the end of this article.
"It may seem surprising that drug users respond to incentives in a sophisticated way. One may think that drug users are poor decision-makers or that addiction makes rational choices impossible. Addiction surely clouds judgement and makes policy in this area difficult, but there is substantial evidence that even drug users respond to incentives."
"Naloxone’s effects may depend on the availability of local drug treatment: when treatment is available to people who need help overcoming their addiction, broad naloxone access results in more beneficial effects."
"Our findings do not necessarily imply that we should stop making naloxone available to individuals suffering from opioid addiction, or those who are at risk of overdose."
Key takeaways on naloxone access and opioid addiction
The authors’ findings sharply diverge with extensive prior evidence that repeatedly suggest a mortality benefit associated with naloxone availability [1,2,3,4,5,6]. Additionally, a similar study by another group of economists, conducted concurrently, showed a significant mortality and societal benefit from broadened naloxone access policies .
After this review was initially written, another very similar study using different data to evaluate opioid overdose mortality and opioid use related to naloxone access laws and Good Samaritan laws found an associated 14 percent decrease in opioid overdose mortality and no increase in use of any opioids associated with naloxone. This further adds to the extensive body of work demonstrating these findings .
Toxicologist, emergency medicine physicians and paramedics, like us, are acutely aware of the opioid epidemic, the complexity of the problem and the challenge to help patients receive effective addiction treatment. Here are seven important takeaways on naloxone use, opioid addiction treatment and the limitations of the Doleac and Mukherjee research paper.
1. Naloxone is not a one-step solution to the opioid epidemic. However, naloxone saves lives.
This paper was written by researchers without medical research experience. This in itself would not be a problem, as economics research is frequently a part of public health considerations. In this case, however, these authors have repeatedly refused to concede the flaws raised by public health experts on this particular research and stand by their conclusions.
With this particular study, the effects of such conclusions could have disastrous consequences for EMS. The findings inappropriately belittle the impact the prehospital response has had on combating the opioid epidemic.
Naloxone does not treat addiction. However, naloxone is the only reversal agent for opioid overdose death in the community and in other situations where ventilatory support is not available. The alternative to the unresponsive, apneic opioid overdose patient is death.
2. Addicts don't seek out naloxone.
While naloxone is a lifesaving antidote, it is also a misery-inducing cause of precipitated withdrawal from opioids. Naloxone is not sought out by patients for a "higher high."
The authors mention "naloxone parties" in the research paper introduction, using as references two speeches by Pennsylvania state representatives who allege to have heard of such parties. This is an irresponsible citation because naloxone parties have been thoroughly debunked in the medical literature and the press at large.
The mention of naloxone parties seems to be presented to bias readers into accepting the conclusions in the paper. Not one single verifiable naloxone party has ever been documented to have occurred. The authors later state that their analysis "confirms these anecdotal reports" of naloxone parties, which, without data on actual naloxone distribution, they have certainly not done.
3. Naloxone availability laws haven't led to increase distribution or utilization.
The associations noted in this study are a result of the continuously evolving epidemic of opioid overdose deaths, which cannot be controlled with the model that the authors use.
Naloxone access laws are used instead of actual naloxone distribution data. Public health policy experts have pointed out that, in addition to some of the state naloxone law data used being incorrect, the intention-to-treat analysis using naloxone laws as proxy for actual naloxone is invalid because these laws have not necessarily led to naloxone distribution [9,10,11,12].
The research paper draws a specific conclusion using naloxone distribution as a variable, without having any actual naloxone distribution data. This falsely states that naloxone distribution laws equal naloxone utilization.
4. There is not fentanyl use data to support research paper conclusions.
Another conclusion this paper draws relates to increased fentanyl use, yet this paper uses "other synthetic narcotics" on crime reports and reported cause of death due to "unspecified drug" as estimates of fentanyl and surrogates for actual fentanyl distribution data. Again, our concern is that without actual data, this paper cannot draw that specific conclusion.
For example, one conclusion states that increased naloxone distribution led to increased fentanyl use, without having real data on either variable. The conclusion also contradicts previous findings [1,2,3,4,5].
5. Naloxone is more beneficial when broader substance abuse services are available.
Naloxone is one of many harm reduction strategies to combat opioid overdose deaths. These strategies remain underutilized while overdose deaths continue to increase. Harm reduction, such as naloxone, cannot treat the medical condition of opioid use disorder, or opioid addiction, but increase survival in order to increase the chances that people are able to recover.
The overwhelming response from public health policy experts and the majority of the medical community is that there remains a major need to increase available treatments in this country, treatments which encompass a variety of different therapeutic options. Furthermore, just as with other medical conditions, like lung disease from smoking, not everyone who receives medical treatment one time is ready or immediately able to continue the process to better health.
6. Stigmatizing addiction exacerbates the epidemic.
This paper uses significant amounts of stigmatic and anachronistic language. Opioid addiction is a polarizing issue, and has adversely impacted a significant amount of our patients and community members. This stigma and stigmatic language are both harmful in terms of the cost of human life, and in the case of this study are polarizing.
7. Significant limitations in data/methodology and stigma lead to faulty assumptions.
There are significant questions being raised in the healthcare community about this study.
First, this paper has not yet been peer-reviewed in the traditional sense, which is not considered an abnormal practice in economics research. Multiple, blinded peer review is a standard in medical research to scrutinize professional conduct [13,14]. This is important to consider when applying non-peer-reviewed research to medical care.
Second, the authors use causal inference to draw their conclusions, a method which is vastly different from those usually presented from randomized controlled trials and standard methods to demonstrate causation used in other usual medical literature.
Third, the authors used Google Search Trends, making the clinical relevance of this data questionable, to assess whether interest in "drug rehab" changed in relation to implementation of statewide naloxone access programs. The result that this data – drug rehab search trends – decreased by 1.4 percent is shown to have a p value of <0.10, which is not the standard used to demonstrate significance in most medical studies.
Finally, "drug rehab," can mean drastically different things (e.g., abstinence only, twelve steps, faith-based, medication assisted therapy, or inpatient or outpatient care), and is not the best way to measure actual treatment for opioid use disorder. Significantly limited availability of many of the better treatment options for opioid use disorder may also have reduced local internet interests.
EMS is uniquely positioned at the interface of opioid-addicted patients and healthcare services, including recovery. Ensuring naloxone is accessible for both community and EMS professional administration is a critical step for resuscitation and recovery. Patients cannot seek treatment for addiction if they are dead. This study bypasses many standard medical prehospital research modalities and makes unsupported conclusions, while continuing to stigmatize the medical disease – opioid use disorder – that our discipline works effortlessly to treat.
The Moral Hazard of Lifesaving Innovations: Naloxone Access, Opioid Abuse, and Crime
1. Clark AK, et al. A systematic review of community opioid overdose prevention and naloxone distribution programs. J Addict Med. 2014 May-Jun; 8(3): 153-63.
2. Heavey SC, et al. 'I have it just in case' - Naloxone access and changes in opioid use behaviours. Int J Drug Policy. 2017 Nov 17;51:27-35.
3. Piper TM, et al. Evaluation of a naloxone distribution and administration program in New York City. Subst Use Misue. 2008; 43(7):858-70.
4. Walley AY, et al. Opioid overdose rates and implementation of overdose education and nasal naloxone distribution in Massachusetts: interrupted time series analysis. BMJ. 2013 Jan 30; 346:f174.
5. Wheeler E, et al. Opioid Overdose Prevention Programs Providing Naloxone to Laypersons-United States, 2014. MMWR Morb Mortal Wkly Rep. 2015 Jun 19; 64(23):631-5
6. Willman MW, et al. Do heroin overdose patients require observation after receiving naloxone? Clin Toxicol (Phila). 2017 Feb;55(2):81-87.
7. Rees DI, Sabia JJ, Argys LM, et al. 2017. "With a little help from my friends: The effects of Naloxone access and Good Samaritan laws on opioid-related deaths." NBER Working Paper No. 23171.
14. McClellan, C. et al. (19 March 2018) Opioid-overdose laws association with opioid use and overdose mortality. https://www.sciencedirect.com/science/article/pii/S0306460318301382
9. Bakhireva LN, et al. Barriers and Facilitators to Dispensing of Intranasal Naloxone by Pharmacists. Subst Abus. 2017 Oct 18:0.
10. Cressman AM, et al. Availability of naloxone in Canadian pharmacies:a population-based survey. CMAJ Open. 2017 Nov 8;5(4):E779-E784
11. Green TC, et al. Perpetuating stigma or reducing risk? Perspectives from naloxone consumers and pharmacists on pharmacy-based naloxone in 2 states. J Am Pharm Assoc (2003). 2017 Mar - Apr;57(2S):S19-S27.
12. Heindel GA, et al. Rising cost of antidotes in the U.S.: cost comparison from 2010 to 2015. Clin Toxicol (Phila). 2017 Jun;55(5):360-363.
13. Ellison G. Is Peer Review in Decline? (July 2011). Economic Inquiry, Vol. 49, Issue 3, pp. 635-657, 2011.
14. AMA Code of Medical Ethics Opinion 9.4.1 https://www.ama-assn.org/delivering-care/peer-review-due-process
About the authors
Ryan Marino, MD, is a medical toxicologist and emergency medicine physician at the University of Pittsburgh, with interest in substance use disorders.
Brian Fullgraf, BS, NRP, is a paramedic with Kirwan Heights EMS in Pittsburgh. Brian has been an active EMS provider and educator since 1997.
Jeremiah Escajeda, MD, is an EMS medical director, prehospital physician and emergency medicine physician at the University of Pittsburgh, interested in EMS and prehospital research.
Continue the discussion and connect with @RyanMarino, @bfullgraf and @jerescajeda on Twitter.