Why Patients Leave: The Role of Stigma and Discrimination in Decisions to Refuse Post-Overdose Treatment

In 2022, an estimated 110,000 people died of an opioid-related drug overdose in the United States (Ahmad et al. 2024) primarily related to illicit fentanyl. However, fatal overdoses comprise only a portion of all overdoses in any given year, since the majority of overdoses are non-fatal (i.e., the individual survives the overdose). Annually, emergency departments (EDs) manage roughly 600,000 non-fatal overdoses once the individual is transported to the hospital (“Non-Fatal Overdoses Short Report” 2023). Tens of thousands of opioid overdoses are reversed and managed by lay bystanders without the need to transport to the hospital, thanks to more widespread access to naloxone, the opioid overdose reversal medication (Dasgupta, Doe-Simkins, and Wheeler 2021).

As the authors of “Revive and Refuse: Capacity, Autonomy, and Refusal of Care After Opioid Overdose” explain, despite hundreds of thousands of individuals being transported to EDs annually following rescue with naloxone, many refuse an evaluation by medical staff and choose to leave the ED before observation for any adverse effects or medical complications. They argue that this presents a challenge for overdose prevention since these patients are at increased risk of rebound-opioid overdose after the naloxone wears off. However, research has found that the vast majority of people are not at risk of rebound toxicity, overdose, or pulmonary edema (Clemency et al. 2018; Farkas et al. 2019; Willman et al. 2017).

Marshall, Derse, Wiener, and Joseph state that the “proportion of patients who make the seemingly irrational choice to refuse any observation is extremely high.” Since the authors view the choice to refuse care as irrational, they argue that it presents a dilemma for ED providers since patients may make these decisions “non-autonomously.” They contend that some patients may have diminished decision-making capacity (DMC) due to their opioid use disorder (OUD), which is characterized by “a disease process which may fundamentally distort patients’ perceptions of risk, and undermine or at least alter their capacity for rational choice.” Most people who overdose, whether fatal or non-fatal, do not meet the criteria for a substance use disorder (SUD) (Schlag 2020).

We disagree with this fundamental premise and assert that people who use drugs (PWUD), even after an overdose, have the capacity to make choices that preserve their own autonomy and dignity, and that decisions to refuse ED care and self discharge are rational responses. The assertion that PWUD cannot make rational choices is based, in part, by the authors’ reliance on the brain disease model of addiction, which fails to pass critical scientific muster (Hammer et al. 2013; Lie et al. 2022; Room 2021). The traditional narrative about people who use drugs implies that, due to fundamental defects in brain chemistry, the ability to make rational decisions is either eliminated or severely diminished (Volkow, Koob, and McLellan 2016). The dopamine hypothesis—from which the disease and hijacked brain models stems—relies on research of the past that fails to meet key validity checkpoints (Hadaway et al. 1979). This claim makes assumptions about how the brain operates that utilize broad enough sweeps to make any deeply learned behavior qualify as a disease (Levy 2013; Lewis 2017). The endeavor of science requires parsimony, and the disease model fails this test on multiple counts (Borowski 2012). To begin, the research was conducted primarily on rats—which are animals that lack a significant prefrontal cortex, the portion of the brain responsible for executive function—and in settings and dosages non-analogous to recreational human drug use (Hart 2014). The claim that dopamine is a catch-all explaining addiction is unsubstantiated, as was the case with the discoveries of other neurons (Berridge 2007). Geneticists and scientists who study the brain are increasingly cautious when tying complex human behaviors to individual neurons or brain regions in light of updated information about the brain (Nutt et al. 2015). For every megapixel of the most advanced brain scanning technology, there are over 850 neurons (Sanders 2023). Claims about how these play out in basic human behavior are less than settled, let alone explaining the complexity of bio, social, racial, and environmental factors that affect criminalized and stigmatized people.

The cultural implications of viewing addiction as a brain disease are equally concerning. Research shows that people who believe addiction is a disease are less likely to seek treatment and more likely to return to use after stopping (Burnette et al. 2019; Miller et al. 1996). Moreover, the general public is more likely to avoid someone they believe to have an addictive brain disease and will carry less optimism about prospects for change (Weisberg et al. 2008). Of additional concern is the way that the criminal legal system relies heavily on the disease meme, in combination with Progressive Era Parens Patriae reforms, to justify forced treatment while extending the reach of drug enforcement through longer probation sentences and deeper surveillance (Tiger 2012). In the eyes of the court, a person who uses drugs is both “sick” and “bad.” The practice of medicine ought to do better than this.

Marshall, Derse, Wiener, and Joseph do acknowledge that “difficulties may be further exacerbated by mistrust, perceived stigma among patients with opioid-related problems, and fear of law enforcement.” In contrast to the authors’ assertion that refusal of post-overdose support in EDs is a presentation of diminished decision-making, this refusal is best understood as a rational reaction to stigma and criminalization. Due to stigma in healthcare—and potential compounding factors like lack of insurance, housing instability, discrimination, and racism—PWUD are less likely to access routine, preventative care (Biancarelli et al. 2019). PWUD make disproportionate use of emergency services, in part because past negative experiences often make them reluctant to seek care until a problem is too acute to ignore. Research reveals that clinicians view PWUD, especially people who inject drugs, negatively and dismissively (Muncan et al. 2020; Surratt et al. 2021; van Boekel et al. 2013). Being disbelieved, ignored, or facing outright contempt push many to avoid or discontinue engaging with healthcare altogether (Carusone et al. 2019; Muncan et al. 2020; van Boekel et al. 2013). Previous or anticipated experiences of discrimination from providers could explain why so few elect post-overdose support in ED settings.

PWUD also face heightened surveillance, coercion, and punishment within healthcare. Common practices—including the use of involuntary drug testing, health record documentation, mandated reporting, and involuntary holds and treatment—are often assumed to improve patient care and health outcomes. In reality, the benefits of these practices are unsubstantiated, and along with police presence in some hospitals, can have serious lifelong impacts like arrest, loss of custody, and deportation (Cohen et al. 2022). Fearing punitive interventions, many PWUD have good reason for opting to avoid healthcare services (Livingston 2020).

In addition to fearing negative provider attitudes and criminalization, PWUD often face hospital restrictions and do not receive appropriate overdose and withdrawal support, harm reduction, treatment, or pain alleviation in EDs (Simon, Snow, and Wakeman 2020). We agree with the authors that ED visits after non-fatal opioid overdoses are potentially a lifesaving touchpoint for patients with OUD to learn about or initiate medications for opioid use disorder (MOUD), such as methadone and buprenorphine that can cut the risk of a fatal opioid overdose by 60% (Larochelle et al. 2018). This point of access could also help to alleviate the large gap between the number of patients who meet the criteria for OUD and those who receive MOUD. However, clinicians may be ill equipped to provide accurate information, care, or MOUD to PWUD due to personal bias and stigma. Few, if any, hospitals prescribe methadone for MOUD, the gold standard for an OUD diagnosis. Many providers also lack training in evidence-based care for PWUD or face financial and bureaucratic barriers to MOUD initiation (D’Onofrio, McCormack, and Hawk 2018; Thomas et al. 2022). Research shows that some providers make PWUD wait longer for care, give inadequate care, or refuse to offer services altogether (Livingston 2020). Compared to people not labeled as drug users, PWUD report inferior care—both related and unrelated to drug use—and are often unable to access pain medication and pain management support (Carusone et al. 2019; Fraser et al. 2017; Muncan et al. 2020).

Marshall, Derse, Wiener, and Joseph argue for an approach in between two poles of forcing a patient to stay and allowing them to leave against medical advice. We assert that post-overdose monitoring for a prolonged period of observation is either unnecessary or does not attend to the post-overdose or withdrawal needs of PWUD. Within the ED, providers can implement patient-centered care through reducing surveillance, coercion, and criminalization and by listening and responding to the needs of their patients (Cohen et al. 2022; D’Onofrio, McCormack, and Hawk 2018). However, given current limitations, we do not believe that EDs are the best place to meet the needs of PWUD post-overdose. Below, we propose several alternatives that ensure the health and wellbeing of PWUD, uphold their dignity and autonomous decision making, and work to end the overdose crisis.

In contrast to many EDs, harm reduction settings like overdose prevention centers (OPCs, also called supervised consumption sites or safe injection facilities) and syringe service programs (SSPs) are often better equipped to respond to the drug-related health needs of PWUD. Evidence from OPCs shows that most people require minimal monitoring after an overdose, and that OPC usage is associated with reduced ED visits and hospitalizations (Harocopos et al. 2022; Lambdin et al. 2022). OPCs and SSPs are places that facilitate safer use practices. They also serve as key points of social connection; provide mental health counseling, medical testing and treatment, and holistic care; and offer meals, showers, and places to rest (Harocopos et al. 2022; Kennedy, Karamouzian, and Kerr 2017; Kerman et al. 2020; Oudshoorn et al. 2021). Importantly, PWUD report that providers in harm reduction settings are more respectful, non-judgmental, and accommodating, leading PWUD to voluntarily continue accessing care in these settings (Muncan et al. 2020). Given the evidence on both stigma in traditional healthcare settings and the lack of stigma in harm reduction settings, we must prioritize developing and strengthening community models of care for PWUD.

At the policy level—beyond harm reduction and ED settings—local, state, and federal governments should work to implement models of true health and safety rooted in community care. Police should not respond to overdose events or drug-related crises; rather, policymakers should implement crisis response teams that are independent of law enforcement and operate without coercion or punishment (El-Sabawi and Carroll 2021). When properly resourced and trained, most harm reduction programs and crisis response teams can respond to overdoses without police and without a trip to the ED. Policymakers must also prioritize decriminalizing drugs to reduce the health, social, and economic harms of criminalization. Coupled with decriminalization, policymakers must invest in community-driven harm reduction, affordable housing, and healthcare.

It is well established that drug prohibition has led to a dangerous drug supply, contributing to staggering overdose rates (Ivsins et al. 2020). In order to truly end the overdose crisis, PWUD must have access to a safe, predictable supply of drugs outside of the unregulated drug market. Safe supply programs must be accessible, affordable, community-based, and consumer-centered. Emerging research on safe supply interventions finds that they not only reduce fatal and non-fatal overdoses but can also yield other positive health, social, and financial outcomes (Kalicum et al. 2024).

The argument that PWUD are not rational actors has been used to justify their mistreatment and to erode their dignity and sovereignty—not just in EDs but in other service systems as well. The increasing focus on mandated or coerced SUD treatment, for example, is rooted in the same set of assumptions that “the problem” lies with PWUD rather than within the systems that purport to serve them. Instead of looking at the limitations of these systems, the pervasive stigma and misformation that inform them, and the mistreatment that PWUD have faced within them, many people chose instead to believe that PWUD are not acting rationally. But there is nothing irrational about refusing care in a setting that is ill equipped to meet your needs, is not needed, has likely mistreated you in the past, and may, in fact, lead to further stigmatization and criminalization, if not death.

We have offered solutions at the programmatic and policy levels that provide care alternatives to reduce stigma and improve health outcomes, all while centering the humanity and self-determination of PWUD. People want to seek care when it is free from stigma, coercion, and criminalization and when it meets their needs. And while we must continue to improve the care of individuals, it is equally important that we acknowledge that most people are dying from a toxic, unregulated drug supply fueled by the war on drugs. We are facing unprecedented loss of life from overdose. This is a crisis that requires an emergency response, and solutions—like a safe, regulated drug supply—that keep people from overdosing in the first place.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.