About this Issue

The opioid epidemic caught America by surprise. Although patterns of problematic substance use often seem to follow the boom-and-bust cycle of infectious epidemics, this one was unusual in several respects. In particular, it struck rural America too, and it did so especially hard. Making matters worse, the opioid epidemic interacted with the legal prescription drug market, leading to bogus prescriptions and the diversion of millions of doses of painkillers. But opioids are in many cases the only effective treatments for pain, so caution is more than usually in order. Addiction is certainly a problem, but so is the under-treatment of pain.

What can we learn about addiction and dependence from the ongoing crisis? Our lead essay this month is by Travis N. Rieder, a Johns Hopkins University bioethicist and author of the recent book In Pain: A Bioethicist’s Struggle with Opioids, which draws on his own experience with opioid dependency.

Responding to Rieder will be Prof. Gail D’Onofrio, MD, who chairs the Department of Emergency Medicine at Yale University; Regina LaBelle, a Distinguished Scholar and Program Director of the Addiction and Public Policy Initiative at the O’Neill Institute for National and Global Health Law at Georgetown University; and Cato Institute Senior Fellow Dr. Jeffrey A. Singer, MD. Each will write an essay, and conversation will continue through the end of the month. Comments will be open for one month as well, and readers are invited to join the discussion.

Lead Essay

Pain, Opioids, and Addiction: Searching for a Nuanced Solution

Pain is a philosophically strange phenomenon. It’s our constant companion, from the stubbed toe, to athletic injuries, to illness, to more mysterious chronic pain. And we tend to act like we understand it—communicating it to our doctors, asking for them to make it go away, and often believing that such a thing is possible. That seeming mundanity, however, hides a mystery: we can talk about pain with our doctors all we want, but we will never make them understand our pain. Pain is an essentially private and subjective experience.

Despite being a philosopher by training, and so having spent a career thinking about things like subjectivity, this simple truth had never occurred to me—not, that is, until I had my foot crushed in a motorcycle accident in 2015. The pain I experienced in the aftermath of that accident taught me many things, but among the most important is that the mysteriousness of pain is responsible for significant insult and injury. The subjectivity of pain means that its presence can never be confirmed—there is no “pain marker” to be found with a blood test, nor definitive evidence to be found with imaging—and so physicians need to determine how to respond to unverifiable reports of pain. Now add to this the reality that opioids are both one of our most potent treatments for pain and capable of causing euphoria, addiction, and eventually death. Clinicians are charged with fixing an ailment that they can’t confirm, and they’re given a tool that is both desired and dangerous. The situation is ripe for tension.

Historically, this tension has led to the American medical community’s opinion of opioids swinging violently back and forth between wholehearted embrace and near-complete prohibition.[i] With the advent of morphine, the hypodermic syringe, and heroin in the nineteenth century, the medical community embraced the aggressive use of opioid analgesics. But when that strategy led to the country’s first overdose epidemic, fear of opioids became dominant, slamming the pendulum back and lodging it firmly in prohibitionist views. This reticence stayed put for nearly a half-century, leading to significant suffering of pain patients. Americans eventually realized the callousness of withholding powerful analgesics, though, and over the latter half of the twentieth century, the pendulum slowly began to swing back toward permissive attitudes, picking up steam in the 1990s. But here again, as humane treatment of pain gave way to recklessly flooding the country with prescription opioids, this glut in supply helped to kickstart a raging drug overdose epidemic, and so—just as it did a hundred years before—fear of the medication has crept back into discussion.

We have been seemingly unable, for at least 150 years, to respond to pain with nuance, compassion, and care, opting instead for one or another ham-fisted approach. Which is preferable: prohibition or recklessness? These aren’t great options, and so we need to land somewhere in the middle. And that, basically, is the starting point of my new book, In Pain: A Bioethicist’s Personal Struggle with Opioids. Combining my experience as a pain and opioid therapy patient with my expertise as a bioethicist, In Pain attempts to articulate what responsible opioid use looks like, and what this has to do with today’s drug overdose epidemic.

Responsible Prescribing

My education on pain medicine began not in the academic literature, but in the hospital. After the fifth limb-salvage surgery following my motorcycle accident, I experienced both harms of a swinging pendulum firsthand. In the morning, when I begged for more pain medication, I was treated with suspicion by the ICU attending; she told me brusquely that she had “noted my request for more pain medication,” and that she would “discuss its appropriateness” with her team before leaving me still writhing in pain. This was an astonishingly callous response, given the fact that the bones of my foot had, just a month before, shattered with such force that they ripped an opening in the bottom of my foot. Indeed, it was a callous response regardless of my reason for asking; her tone and curtness made it clear that she didn’t believe my pain was as bad as I claimed, and such disbelief is an insidious form of disrespect.[ii]

Later that day, though, I would eventually get one of the surgical residents to organize a pain management consult, and the pain team that eventually made their way to my room prescribed lots of opioids, plus a non-opioid called gabapentin, and intravenous acetaminophen. Those pain docs worked magic, as far as I could tell, sending me into fairly comfortable oblivion.

Although at the time, it seemed that the latter treatment was the appropriate one—aggressively treating my pain when I asked—that wasn’t the end of the story. I only ever saw the pain doc one more time, on the following day, when he stopped by for 30 seconds to ask if my pain was under control. And neither he, nor anyone else on his team, ever discussed the risks of opioids with me. We didn’t talk about tolerance, dependence, withdrawal, addiction, or overdose. We talked about getting “ahead of the pain” and staying there. This team started a pain management protocol that they never intended to manage over the long term, and so my surgeon picked it up when I left the hospital, signing off on refills and escalating my dose as tolerance reduced the drug’s effectiveness. Worse yet, when my trauma surgeon finally advised me to get off the pills a month later, no one from any of my clinical teams offered guidance or could provide a competent taper.

In the book (and elsewhere) I describe in detail the devastating withdrawal that resulted from this mismanagement, so I won’t do that here. For present purposes, I will simply note the obvious, that this sort of reckless prescribing is irresponsible. In my case, it caused terrible suffering, and I was one of the lucky ones—I eventually succeeded in tapering off the medication. But as I would go on to discover over several years of talking to patients and doctors, this particular failure is not uncommon. Doctors know how to write a prescription, but not many of them know how to actually manage opioids, or when (and for how long) they are appropriate. My experience made clear that, at this moment in history, we are simultaneously erring on both sides of the opioid pendulum: doctors are both under-prescribing, and treating pain patients with suspicion, while also recklessly prescribing, and continuing to risk treatment-originated dependence and addiction.

Many people at this moment are calling for responsible prescribing in a narrow sense. They note our recent history of overprescribing and recognize the need not to write prescriptions when they’re not called for. We’re even getting better at recognizing a more complex version of this sense of responsibility—noting not only that we shouldn’t prescribe Percocet for tooth extractions or moderate injury pain, but that when opioids are called for, we shouldn’t prescribe more than are necessary.[iii] Without an evidence base, we’ve massively overprescribed even for truly serious pain, like that resulting from surgery. But now we’re starting to accumulate that evidence base and improve.

What my work adds to this discussion is the observation that appropriate initiation is not the only relevant aspect of responsible opioid prescribing. Prescribed opioids must also be appropriately managed over the long term, beginning with patient education and counseling and continuing through a humane taper off the medication. Responsible pain medicine, in other words, must be nuanced. No ham-fisted policy of the “one size fits all” variety will fix opioid prescribing, because responsible pain medicine isn’t about limiting access to opioids or, conversely, ensuring that every last pain patient has access to opioids. That way lies the swinging pendulum, and it cannot solve this problem.

Stopping the swing, however, will be hard. Responsible opioid prescribing will require that every clinician who treats pain knows when opioids are called for, how much and for how long they are called for, how to counsel and educate patients, how to manage the medication over the long term, and how to taper dependent patients off the medication. They also need to have the relationships (and willingness) to partner with experts in pain or addiction medicine when particular patients outstrip their expertise. Somewhere between none and very few of these expectations are currently in place for most clinicians, and endorsing them will require major changes to the healthcare system.

Fixing Pain Medicine Won’t Solve the Drug Overdose Crisis

Clinicians can get better at prescribing opioids. A seductive narrative suggests that if they do this, it will solve the opioid epidemic. Unfortunately, that’s not true. As a matter of fact, not only is it untrue, but the simplistic notion that there is “an opioid epidemic” that is synonymous with the prescription opioid problem is responsible for a lot of harm—both to people with opioid use disorder and to pain patients.

Prescription opioids almost certainly contributed to today’s drug overdose crisis. When careless overprescribing really kicked off, in the 1990s, it led to a striking uptick in overdose death. Between 1999 and 2010, prescriptions for opioids quadrupled, which was matched by a quadrupling of the prescription opioid overdose death rate. The trend lines are hard to ignore.

That was not, however, the end of the story. Singling out this drug supply as the problem led to a supply-focused response: restrict prescribing. After all, if loose prescribing kills people, then restrictive prescribing should prevent those deaths. Right?

That is not, however, what happened. Although restricting the supply of pharmaceutical opioids did bend the curve down on prescription opioid overdose death, it did not solve the opioid crisis; in fact, since 2012, when prescription opioid overdose deaths began to decline, increase in the overall opioid overdose death rate sped up. This is because as soon as access to prescription opioids was restricted, the death rate from heroin increased dramatically, followed soon after by an absolute skyrocketing of the death toll from synthetic opioids like fentanyl. What seems to have happened is that at least some of the people who were taking prescription opioids were willing to transition to heroin when their supply was cut off; and as heroin became more and more contaminated with fentanyl and its analogues, we’ve been brought to the catastrophic heights of America’s current overdose crisis.

This telling of the history suggests that trying to solve the opioid epidemic by focusing on the supply of prescription opioids harmed a particular population—those with opioid use disorder—by either forcing or incentivizing a switch from their safer supply of pharmaceutical grade opioids to an increasingly contaminated supply of heroin. But there is another group that has been harmed by this policy focus, and that group is made up of pain patients who don’t have an opioid use disorder. Some of these patients are harmed by simply being underprescribed (since opioids do have some proper indications), while others are harmed by being abandoned or forcibly tapered, both of which can precipitate terrible withdrawal.

This last group of patients, who are sometimes called “legacy patients” (as they are a legacy of past, aggressive prescribing) or “orphan patients” (if they are abandoned by their prescribing clinician and cannot find anyone to fill their prescriptions), can be on truly massive doses of opioids. Because clinicians have been aggressively prescribing opioids for decades, and opioids cause tolerance, some patients are on many hundreds of milligrams of morphine or its equivalent, and the prospect of withdrawal can be terrifying.[iv] As someone who has experienced withdrawal, I completely understand this. Withdrawal can range from uncomfortable to excruciating, and at its most severe can drive one to contemplate suicide. Forcing someone into that position because doctors have changed their minds on whether they endorse some therapy seems cruel in the extreme.

Perhaps the most controversial claim I make in my book, then, is that clinicians should not unilaterally taper legacy patients without their consent. There was real overprescribing in the past, and we should correct much of it. But reducing opioid supply by forcing patients into withdrawal is not an appropriate response to America’s drug overdose crisis. We eventually must stop compounding the harms of our past behavior.

Moving beyond Supply

The broad lesson of In Pain is that although American pain medicine is desperately broken, fixing it won’t solve the opioid epidemic. Acting as if it will is likely to lead to un-nuanced policy that hurts both pain patients and those suffering from addiction.

So what will solve the broader drug overdose crisis? Well, moving beyond supply-focused responses can help us to realize what sorts of interventions are absolutely required. America has a population of people already addicted to various drugs, including opioids, and our goal should be reducing the morbidity and mortality that goes along with that addiction. The above exploration of the failures of supply-focused interventions regarding prescription opioids helps to explain why America’s War on Drugs has been such an abject failure: when a person’s health condition is defined by compulsive behavior despite harmful consequences, introducing ever-more harmful consequences is a bad way to try to change behavior. Some people will continue to be willing to engage in the behavior, and so the primary result of the intervention is escalating harm.[v]

Rather than engage in policy that exacerbates the harms of drug use, then, we should look for interventions that reduce those harms. These include, at a minimum, interventions like syringe-exchange programs (an evidence-based strategy for reducing the disease burden among people who inject drugs) and distribution of the overdose-reversal medication naloxone; more ambitiously, I think moving beyond supply should lead us to open safe-injection sites (sometimes called overdose prevention sites), where people who use drugs can do so in the presence of healthcare practitioners, with sterile supplies and naloxone on hand. Most radically, perhaps it should lead us to consider a safe-supply policy, on which doctors would be allowed to prescribe pharmaceutical grade heroin or hydromorphone for people with opioid use disorder who don’t want to risk the contamination of street drugs, but aren’t ready to seek treatment.

This last point about safe supply makes clear the extent to which harm reduction is a theme in my views on drugs, both prescription and illicit. Whether a patient is on high-dose oxycodone and fearful of withdrawal, or using illicit heroin and living with active addiction, I don’t think taking away their supply is an ethical response. In either case, restricting supply can be devastatingly harmful, either through directly causing withdrawal and recurrence of whatever pain they were medicating, or by driving them to less safe options.

I don’t expect, in these few words, to have convinced anyone who is vehemently opposed to harm reduction. The arguments canvassed here—laid out in more detail in In Pain—are largely about responsible opioid prescribing against the backdrop of the drug overdose crisis. Pain medicine and the broader crisis are linked, however, through opioids, and it seems to me that the lessons learned from investigating how we should view prescription opioids can provide insight into how we view drugs and addiction writ large.


[i] This paragraph is a summary of Chapter 3 of In Pain: A Bioethicist’s Personal Struggle with Opioids. For detailed sources on the historical narrative told here, please consult that chapter.

[ii] And of course: it should be noted that I’m privileged to have experienced this sort of epistemic injustice only rarely in my life. As a white man, I am actually more likely to have my claims of pain taken seriously than many others. An important component of the harm of not being believed, then, is that it furthers various forms of injustice.

[iii] A leader in this area is the organization Michigan OPEN (the Opioid Prescribing Engagement Network), which is both generating an evidence base for how many pills given pains and procedures tend to need, as well as providing resources for clinicians and patients. More information on all of this can be found at their website, here: https://michigan-open.org.

[iv] For context: the CDC recommends caution when escalating patients past 50 morphine milligram equivalents (MME), and advises against escalating past 90 MME without careful justification. I was on a total of about 170 MME when I began my too-aggressive taper, and I was immediately cast into terrible, acute withdrawal. Legacy patients can be on many hundreds, or even 1000 or more MME due to years of tolerance.

[v] This is only one explanation of why the War on Drugs failed. It is also morally problematic for other reasons, like its racist implementation.

Response Essays

Prescribers and Policymakers Must Both Get Better at Harm Reduction

I have been a practicing general surgeon for more than 30 years. A major part of my work involves the treatment of acute and chronic pain. Sometimes that pain is a direct result of the operations I perform. Travis N. Rieder reminds us, through his personal experience, that pain is personal, individualized, and often horrific and insufferable. He rightly points out that opioids can be very effective pain relievers, making life bearable again for many who are immobilized and demoralized by pain.

Clinicians have varying philosophies on the treatment of many conditions. Just as some doctors may overprescribe antibiotics or overtreat hypertension or diabetes, others may display opposite dispositions. These differences also apply to pain management. The subjective nature of pain makes it very difficult to establish objective evidence-based standards to resolve clinicians’ management differences. One-size-fits-all approaches are inappropriate and harmful to patients. Rieder’s encounter with pain acutely illustrates this problem. Even more compelling to this doctor, Rieder brings attention to a major shortcoming of most medical training programs regarding the treatment of pain: physicians are trained to prescribe opioids and other medications to treat pain, but they are rarely taught how to properly manage the withdrawal from this type of drug that, like many others, can induce physical dependency and tolerance.

Physicians who prescribe beta-blockers, anti-depressants, anti-epileptics, and other drugs known to induce dependency are trained to supervise safe, gradual tapering to avoid withdrawal complications. They advise their patients against abrupt cessation of their medication without first seeking professional guidance. As Travis Rieder painfully learned, this is rarely what happens when it comes to prescribing opioids to manage pain. Most of us physicians know what to administer to a patient in pain, but few of us know how to safely and compassionately wean patients off the pain medication. If a patient seeks advice on how to safely taper off the drug, the best most physicians usually offer is an educated guess. This failing put Rieder, and no doubt puts countless other patients, through an avoidable but agonizing struggle to shake off dependency.

It is irresponsible to place a patient on a drug without being aware of how to safely get the patient off the drug. It falls to medical professionals and educators to rectify this deficiency. From a clinician’s perspective, this precept of harm reduction is a crucial lesson of Rieder’s essay.

Many people confuse addiction and dependence, and this unfortunately suffuses much of the policy debate surrounding the prescribing of opioids. The American Society of Addiction Medicine defines addiction as a “chronic disease of brain reward, motivation, memory and related circuitry…characterized by the inability to consistently abstain, impairment in behavioral control, [and] craving” that continues despite resulting destruction of relationships, economic conditions, and health. Addiction has a biopsychosocial basis with a genetic predisposition and involves neurotransmitters and interactions within reward centers of the brain. Some experts believe addiction is a learning disorder in which behavioral patterns are automatized as mechanisms for coping with stress or trauma. A major feature of addiction is compulsiveness. This compulsiveness is why alcoholics or other drug addicts will return to their substance of abuse even after they have been “detoxed” and despite the fact that they know it will further damage their lives.

Addiction is not the same as dependence. Physical dependence refers to the physiological adaptation to the drug such that abrupt cessation or tapering off too rapidly can precipitate a withdrawal syndrome, which in some cases can be life-threatening. Tolerance is an aspect of physiological adaptation, in which increasing dose of a medication become necessary to achieve the desired effect. When a patient who does not suffer from addiction is properly tapered off of the drug on which they have become physically dependent, they do not feel a craving or compulsion to return to the drug.

The lack of insight into the often subtle distinctions between dependency, tolerance, and addiction has led to what Rieder calls “ham-fisted” policies that impose one-size-fits all constraints on opioid prescribing to patients in both acute and chronic pain. I agree. In hasty reaction to the opioid-related overdose crisis that mobilized national attention earlier this century, many states have memorialized in official policy the opioid prescribing guidelines published by the U.S. Centers for Disease Control and Prevention in 2016. These guidelines have come under much criticism from scholars and clinicians for lacking a strong basis in the evidence. Limits have been placed on the number as well as the “morphine milligram equivalent daily dosage” (MEDD) of opioids that may be prescribed. The MEDD metric has come under severe criticism for its inaccuracy and inappropriateness.

The almost hysterical reaction of policymakers to the overdose crisis propelled a reversion to pre-modern approaches to the treatment of pain. Many patients in severe acute or chronic pain are now, once again, being undertreated for their pain, as they were prior to the late 1980s. Many chronic pain patients whose pain has been controlled with chronic opioid use are being rapidly and recklessly tapered off their medication, putting them through the combined agonies of resumption of pain and protracted withdrawal, often rendering them disabled prisoners of their condition. Some, in desperation, turn to the dangerous black market for relief. Others resort to suicide.

Complaints by academicians, clinicians, and the American Medical Association (Resolution 235) finally caused the CDC to issue a clarification in April 2019, noting, “Some policies, practices attributed to the Guideline are inconsistent with its recommendations.” Among the misapplications of the guidelines it noted were those that result in “hard limits or ‘cutting off’ opioids,” stating the “Guideline does not support abrupt tapering or sudden discontinuation of opioids.” Yet the statutory and regulatory restrictions remain unchanged.

Much of this ham-fisted policy is driven by the mistaken belief that the overprescribing of opioids for pain is what caused the overdose crisis. Numerous studies in the 1970s, 1980s, and 1990s led to the correct conclusion that patients were being undertreated for pain based upon the fear of risk of overdose or addiction. As clinicians and patients were encouraged to overcome such fears and more liberally treat pain, the volume of prescription opioids increased dramatically, particularly from 1999-2012. During this same period opioid-related overdose deaths quadrupled. It is easy to understand the temptation to draw a connection.

However, research using data provided by the National Survey on Drug Use and Health and the U.S. Centers for Disease Control and Prevention clearly shows no correlation between prescription volume per capita and “past month non-medical use” as well as “pain reliever use disorder in the past year” in persons aged 12 and over. The lack of correlation was recently confirmed in Germany, which has the second highest opioid prescription volume in the developed world. Moreover, data from the NSDUH show less than 25 percent of non-medical users of prescription pain pills ever obtain them from a prescription.

Studies also show the overdose potential of opioids, when used as directed in the medical setting, range from 0.022 percent to 0.04 percent. Multiple Cochrane systematic reviews point to extremely low addiction rates in chronic noncancer pain patients receiving long-term opioid therapy, and a recent report found a total “misuse” rate of 0.6 percent among more than 568,000 patients receiving opioids for acute post-surgical pain during an eight year period.

Beginning around 2010, government policies succeeded in dramatically reducing prescription volume. These policies included Prescription Drug Monitoring Programs that nudged or even frightened clinicians into prescribing fewer opioids; the promotion of abuse-deterrent formulations of opioids; and mass arrests of corrupt clinicians operating “pill mills” supplying non-medical users with their drug of choice. Prescription volume peaked in 2012. Yet as total prescription volume dropped by 29 percent between 2010 and 2017 the opioid-related overdose death rate continued to rise, increasing 16 percent from 2016 to 2017. Also changing between 2010 and 2017 were the specific drugs that caused the opioid-related deaths. The proportion of opioid-related deaths attributable to heroin and fentanyl has dramatically risen while those due to prescription opioids have fallen off. For example, in 2017, CDC data reveal that fentanyl or heroin were involved in 75 percent of opioid-related overdose deaths, up from 28 percent in 2010. In 2017 just 30 percent of opioid-related overdose deaths involved prescription opioids, down from 52 percent in 2010, and 40 percent of deaths involving prescription opioids also involved heroin or fentanyl. These trends suggest non-medical users are migrating from diverted prescription opioids to heroin and illicitly produced fentanyl as legally produced opioids become more expensive and difficult to obtain in the underground market. The evidence indicates the current emphasis on reducing prescription opioid volume only serves to worsen the situation by driving non-medical users to more dangerous and deadly opioids in the black market.

Research from the University of Pittsburgh reveals the overdose rate from the non-medical use of licit and illicit substances has been on a steady, exponential increase at least since the late 1970s, and it is showing no evidence of deviating from that trend. The only thing that has changed over the years is the particular drug in popular use and responsible for those deaths at any given time. In the early part of this century the popular drugs for non-medical users were diverted prescription opioids. Next it became heroin. For the past several years it has been heroin and fentanyl. Another worrisome trend is that non-medical users are taking risks that previous generations were less willing to take. For example, Cicero, et al found 33.3 percent of heroin addicts entering rehab in 2015 initiated non-medical drug use with heroin, as opposed to 8.7 percent 10 years earlier.

The evidence shows the overdose crisis had its beginnings well before prescription opioids began to gain prominence as a cause of death. Contrary to the narrative that drives prevailing policy, the evidence suggests that the overdose crisis from the non-medical use of licit and illicit drugs has sociocultural and psychosocial etiologies and is exacerbated by the dangers inherent in accessing and using drugs in an underground market fueled by drug prohibition.

As mentioned above, Germany has the second highest opioid prescription volume in the developed world. It’s pattern of opioid prescribing mirrors that of the United States, surging from the mid-1990s and then peaking in 2012, after which it receded. Yet Germany’s opioid-related overdose rate has consistently been among the lowest in the developed world. One major reason is that Germany has widely adopted harm reduction strategies since the 1980s. These include needle exchange, safe consumption sites, and “safe supply” programs as well as Medication Assisted Treatment, heroin assisted treatment, and other rehabilitation programs. Rieder wants policymakers to move to harm reduction strategies to address the overdose crisis in the United States, and I couldn’t agree more.

Rieder is correct when he writes “the war on drugs has been an abject failure.” It has also been the principal force behind the overdose crisis. Furthermore, engaging the criminal justice system to deal with substance use disorder is an irrational way to treat a behavioral disorder that is defined as compulsive behavior despite negative and self-destructive consequences. Punishment won’t work to cure a disorder for which, by definition, punishment doesn’t work.

Doctors prescribing opioids in the medical setting, and policymakers addressing their use in medical and non-medical settings, must better appreciate the complexity and nuance surrounding use of these helpful and powerful drugs. Both settings have in common the need for larger doses of harm reduction.

Fight the Opioid Epidemic with Stigma-Free Treatment

Travis Rieder’s essay is hard to read from a physician’s perspective as it highlights multiple times when the health care system failed him. We can do better.

The pain trajectory to opioid misuse, addiction, and overdose is unfortunately real and tragic.

Long term opioid use often begins with the treatment of acute pain. Shah and colleagues[1] studied a random sample of commercially insured, opioid naive individuals who received an opioid prescription. They found that with one day of use, the probability of use at one year was only 6%, but it rapidly increased to 13.5% for those prescribed for > 8 days, and 29.9% with > 31 days of use. Thus, educating physicians about safe prescribing and supplying the correct amount of pain reliever depending on the injury or illness is one of many steps that we need to do to save lives. However, chronic pain is highly prevalent in the U.S. population, affecting over 100 million people, or one in three individuals, with 25 million Americans suffering from daily pain.[2] It is estimated that 10% of patients with chronic pain misuse opioids.[3] Consequently, the NIH is investing substantial dollars in developing new medications and treatments for pain; the Center for Disease Control has offered guidelines for safe prescribing; and the number of opioid prescriptions has decreased.

There is enough blame to go around for the creation of the perfect storm.

Physicians were being told in the mid 1990s that they undertreated pain, and they became more than willing to prescribe more pain medications with little scientific evidence. The American Pain Society and other advocacy groups lobbied for pain to be included as the fifth vital sign, along with temperature, pulse, respiration, and blood pressure. Purdue Pharma aggressively marketed the longer acting sustained release opioid formulation OxyContin with few facts related to its addictive potential, and it targeted primary care practices and areas of high prevalence of pain, such as the rural Appalachian Valley. Regulatory bodies, such as the Joint Commission for Hospital Accreditation, released new pain standards in 2001, and finally in 2006 CMS linked reimbursement to patient satisfaction with pain management.

In retrospect it was the perfect storm. However, today the overdose rates from prescription opioids are decreasing, and more and more deaths are attributed to heroin and increasingly use of synthetic opioids such as fentanyl. This is even more frightening, as synthetic fentanyl is cheap, easily purchased through the mail, often sold and marketed as other illicit substances such as heroin or cocaine, and it’s strikingly more lethal. In 2017 synthetic opioids have become twice as common in overdose deaths as prescription opioids or heroin.[4] Thus, we need to learn from our mistakes and develop prevention, treatment, and harm reduction strategies partnering with law enforcement and public health.

Stigma is a large factor preventing access to treatment.

Once we all—including health professionals, the public and the judicial system—understand that opioid addiction is a chronic, relapsing disease that can be treated with proven medications and is not a moral failure, we will be closer to saving lives. Opioid agonist treatment with methadone or buprenorphine does work and has proven to reduce illicit use, craving, HIV and hepatitis C transmission, crime, and contact with the judicial system; it also increases retention in treatment and improves quality of life. So why is it that so many patients refuse treatment, and why do physicians fail to treat? First, the stigma associated with opioid addiction is profound both in the lay population and the health care field, and it is one of the major reasons individuals with OUD do not seek treatment. The lack of knowledge surrounding addiction is pervasive and disheartening. Words matter,[5] and individuals are often labeled as “addicts” instead of individuals with a disease, opioid use disorder (OUD); their urines are termed “dirty or clean” instead of positive or negative; care is often terminated if the individual “fails treatment” instead of returns to use, where treatment should be escalated, not withdrawn.

An individual with diabetes would never have their care terminated if they presented one or many times with high blood sugars. No one would contemplate not offering another prescription for insulin. But this is exactly what we do with OUD. While diabetes has many behavioral treatments as part of the care, including weight loss, dietary restriction, and nutritional guidelines, we know that the treatment is medication (insulin or oral preparations). With OUD the term medication assisted treatment (MAT) continues to be used instead of the more accurate terms of opioid agonist treatment (OAT) or medications for opioid use disorder (MOUD). Countless studies[6] and reports[7] have emphasized that medications are as effective with or without counseling and should never be withheld, but we continue perpetuating this falsehood that medication is merely optional.

Many individuals with OUD end up in the judicial system, which lacks the will and ability to treat the underlying disease. In Connecticut alone, there were approximately 1,000 deaths in 2017 from overdose, and 55% of these individuals had been previously incarcerated in a Connecticut prison or jail.[8] Yet few departments of correction offer methadone or buprenorphine. A National Academy of Medicine report states that “withholding or failing to have availability of all classes of FDA-approved MOUD in any health care justice setting is denying appropriate medical treatment.7

Physicians, patients, and family members often think we are replacing one opioid for another, or one addiction for another. This is simply not true. MOUD are prescribed/administered specifically by physicians. The hallmarks of addiction include craving, consequences, and lack of control. When an individual receives MOUD they return to work, have meaningful relationships, and no longer engage in destructive activities despite knowledge of their consequences.

Strategies for saving lives.

Families and friends need to encourage their loved ones to seek help. Health professionals must recognize OUD and identify patients in need of assessment, treatment, and referral, using signs that are often obvious. OUD is estimated to be present in 3-17% of primary care practices,[9] and opioid-related emergency department (ED) visits increased 30% from 2016 to 2017.[10] Patient and healthcare system challenges exist and must be overcome. Physicians currently must obtain a Drug Addiction and Treatment Act of 2000 (DATA 2000) waiver, requiring 8 hours of training, to prescribe buprenorphine. This is truly laughable when a physician can prescribe countless opioids for pain without any further training. However, the law is an Act of Congress and most likely will be in place for the foreseeable future. Thus, we should all be required to obtain the training.

At Yale School of Medicine, with help from SAMHSA, as well as several others around the country, this training is being initiated for medical students, physician assistants, as well as to the advanced practice nurses in the School of Nursing. Residents are now receiving training in many primary care specialties including emergency medicine. After all, do we refuse to treat, or are we unprepared to care for, patients with other life-threatening illnesses, such as heart attacks or strokes? Should this preparation be optional? Shouldn’t obtaining a DATA 2000 waiver be a requirement for hospital credentialing?

Hospitals and pharmacists need to stock buprenorphine and have systems in place for patients to fill their prescriptions seven days a week. We know from a study conducted at Yale-New Haven Hospital emergency department that ED-initiated treatment with buprenorphine more than doubles engagement in treatment at 30 days when compared with receiving a referral alone,[11] is cost effective,[12] and saves lives. So why not require this evidence-based practice? The cost of the medication is less than an ED visit, hospitalization, or the staggering increase in heart valve replacements due to endocarditis. The overall economic cost of the U.S. opioid crisis is estimated to be more than $500 billion per year.[13]

Identifying community opioid programs and providers and partnering with them to provide timely referrals and other services is essential. State laws that require extensive counseling that precludes primary care offices form prescribing buprenorphine need to be eliminated. As many individuals have unstable housing, shelters should be monitored and drug-free so that individuals in treatment feel safe using them.

Harm reduction strategies are also imperative. We should be dispensing naloxone to patients with OUD or at risk for OD, and their families or friends. Insurance companies should include naloxone in all plans without copays, and pharmaceutical companies should not be overcharging for this medication or other MOUD.

We need to increase the workforce for addiction treatment.

Sadly, Travis Rieder did not have the benefit of an addiction specialist involved in his care from the beginning. Multiple surgeries require thoughtful and effective pain management. Developing a physical dependence on opioids due to these surgeries does not necessarily mean that one will develop an addiction. Cancer survivors and individuals with sickle cell disease may develop physical dependence, meaning they require more medications due to tolerance, or may develop withdrawal symptoms when the medication is withdrawn, but they do not meet the other criteria for addiction. Having an addiction specialist consulted will aid the physician in appropriately prescribing the opioids, adding other treatments for pain, and recognizing when a patient may have developed an addiction and need treatment.

Addiction Psychiatry has been a subspecialty since 1992 but is limited to physicians who have completed a psychiatry residency. Fortunately, Addiction Medicine has recently been formally recognized as the first clinical multi-specialty subspecialty recognized by the American Board of Medical Specialties (2016) and is administered by the American Board of Preventive Medicine (ABPM). Any physician with primary board certification such as Internal Medicine, Family Medicine, Emergency Medicine, Pediatrics, Preventive Medicine, Obstetrics and Gynecology, Anesthesia, or others can apply through June 2021 for Addiction Medicine subspecialty certification via the ABPM Practice Pathway.

As is true for all new ABMS recognized fields, this pathway will close, and then a twelve-month, full-time Addiction Medicine fellowship will be required after primary residency completion. The Accreditation Council for Graduate Medical Education (ACGME), the entity that accredits all physician training after medical school, recognized the subspecialty in 2018 and then began accreditation of fellowships. ACGME provides the gold standard accreditation for physician training. As of June 2019, there were 54 ACGME accredited addiction medicine fellowships, 20 programs previously accredited by the American College of Academic Addiction Medicine which are transitioning to ACGME accreditation, and another 15 emerging programs in development.

It is anticipated that we need 7,500 full-time addiction medicine specialists by 2025 and into the future; there are currently about 2,000. This will require at least 125 training fellowships with 2-3 fellows graduating from each annually. Policies that can assist in paying hospitals for these fellowships and encourage physicians to take this additional year of training are needed. This could include working in rural or underserved areas and with underserved populations both in rural and in urban areas.

End the opioid crisis with a multifaceted approach.

A multifaceted approach involves partnering with health care providers, public heath, law enforcement, and the judicial system. The stigma of addiction needs to be removed, and individuals need to come forward to tell their story and not be ashamed or afraid of repercussions. Treatment capacity and harm reduction strategies need to be increased so that access to care can be available and affordable in all communities, rural or urban. Patient and system barriers to treatment must be broken down. Lastly the Addiction Medicine workforce must be increased to help improve the health of the public through this epidemic and other escalating substance misuse and disorders.


[1] Shah A, Hayes CJ, and Martin BC. (2017). “Characteristics of Initial Prescription Episodes and Likelihood of Long-Term Opioid Use — United States, 2006–2015.” Morbidity and Mortality Weekly Report (MMWR), 66(10):265-269.

[2] Loeser J, D., (2012). Relieving pain in America Clin. J. Pain,28pp.185-186

[3] Garland, E. L., Froeliger, B., Zeidan, F., Partin, K., & Howard, M. O. (2013). The downward spiral of chronic pain, prescription opioid misuse, and addiction: cognitive, affective, and neuropsychopharmacologic pathways.Neuroscience and biobehavioral reviews,37(10 Pt 2), 2597–2607. doi:10.1016/j.neubiorev.2013.08.006

[4] Hedegaard H, Miniño AM, Warner M.Drug overdose deaths in the United States, 1999–2017. NCHS Data Brief, no 329. Hyattsville, MD: National Center for Health Statistics. 2018.

[5] Botticelli, M. P., & Koh, H. K. (2016). Changing the language of addiction.Jama,316(13), 1361-1362.

[6] Mattick RP, Breen C, Kimberly J, Davoli M. Buprenorphine maintenance versus placebo or methadone maintenance for opioid dependence. Cochrane Database Syst Rev. 2014 Feb 6;(2):CD002207. Doi10.1002/14651858.CD002207.pub4

[7] National Academies of Sciences, Engineering, and Medicine. (2019).Medications for opioid use disorder save lives. National Academies Press.

[8] State of Connecticut. Office of Policy and Management monthly indicators report https://portal.ct.gov/-/media/OPM/CJPPD/CjResearch/MonthlyIndicators/2014-2018/2018MonthlyIndicatorsReport/MonthlyIndicatorsReportSEPT2018DRAFTpdf.pdf?la=en September 2018

[9] Vowles, K.E., M.L. McEntee, P.S. Julnes,et al.2015. Ratesof opioid misuse, abuse, and addiction in chronic pain: a systematic review and data synthesis.Pain156:569–57

[10] Vivolo-Kantor, A. M., Seth, P., Gladden, R. M., Mattson, C. L., Baldwin, G. T., Kite-Powell, A., & Coletta, M. A. (2018). Vital signs: trends in emergency department visits for suspected opioid overdoses—United States, July 2016–September 2017.Morbidity and Mortality Weekly Report,67(9), 279.

[11] D’Onofrio G, O’Connor PG, Pantalon MV, Chawarski MC, Busch SH, Owens PH, Bernstein SL, and Fiellin DA. (2015). “Emergency Department-Initiated Buprenorphine/Naloxone Treatment for Opioid Dependence: A Randomized Clinical Trial.” JAMA, 313(166):1636-1644.

[12] Busch SH, (2017) Fiellin DA, Chawarski MC, Owens PH, Pantalon MC, Hawk K, Bernstein SL, O’Connor PG, D’Onofrio G. Cost Effectiveness of Emergency Department-Initiated Treatment for Opioid Dependence. Addiction 2017, doi: 10.1111/add.13900.

[13] Council on Economic Advisers. The underestimated cost of the opioid crisis. Accessed November 26, 2018.https://www.whitehouse.gov/sites/whitehouse.gov/files/images/The%20Underestimated%20Cost%20of%20the%20Opioid%20Crisis.pdf

Criminalizing Addiction Isn’t Working. Prevention and Treatment Deserve a Chance.

Travis Rieder’s very personal story about his journey and our nation’s fractured response to pain highlights an important narrative in our nation’s health care system. This story also highlights areas of commonality between the management of pain and the treatment of substance use disorders, both complex conditions calling for a nuanced approach.

People with pain and those with a substance use disorder suffer stigma, shame, and lack of access to effective treatments. This is well illustrated in Dr. Rieder’s essay when the ICU attending gives a “callous response” to his request for pain medication, silently judging his request as possibly indicating “drug seeking” behavior. People with substance use disorders are regularly subjected to this type of judgment throughout our health care system and in society. While such judgments are sometimes made silently, they are expressed in stark terms throughout the nation’s legal and criminal justice system. The chronic and relapsing nature of addiction—the very symptoms of addiction—are punished and criminalized. Unfortunately, both people with pain and those with substance use disorder are too often demonized and their conditions dismissed. Patients are sometimes divided into those with real pain symptoms, versus those who may have developed an addiction to opioids and are merely seeking more pain medications. The reality is far more complex. We know from surveys that people may misuse opioids to manage their own pain, and as a result some go on to develop an addiction, making treating their conditions more complicated.

Pain patients deserve more and better options to manage their pain, and abruptly stopping opioid treatment is cruel. As Dr. Rieder points out, major changes to the healthcare system are necessary to stop the swing between overprescribing of opioids and forced tapering. Major changes to the healthcare system are also needed to help the estimated 2 million people with opioid use disorder. A good place to start is by integrating the addiction treatment system into the healthcare system. Since substance use disorders have long been considered moral failings requiring a criminal justice response, the health care system is ill equipped to respond effectively to any type of substance use disorder. As with mental health, our nation’s correctional system has become the de facto treatment system for people with substance use disorders. And every day, people with opioid use disorder are incarcerated, forced into withdrawal, and denied treatment medications that would prevent abrupt withdrawal and give them a better chance of entering recovery and avoiding overdose.[i]

During my tenure as a political appointee in the Office of National Drug Control Policy during the Obama administration, we focused intently on reforming our nation’s addiction treatment system and developing a response to the rising tide of opioid overdose deaths. Decreasing opioid prescribing rates was part of this strategy, but it did not comprise the entirety of the policy response to curb overdose death rates. In 2016, the Obama administration worked with Congress to include over $1 billion in new funding to expand evidence-based treatment for opioid use disorders through the 21st Century Cures Act. This funding, coupled with increasing awareness about the risks associated with opioid misuse, the Affordable Care Act, Medicaid expansion, naloxone distribution, and support for syringe services programs formed the basis for the federal response to opioid morbidity and mortality.

The nation has only recently seen a decrease in prescription opioid involved overdose deaths.[ii] According to CDC statistics, overdose deaths involving prescription opioids leveled off in 2012 and were then followed by increases in overdose deaths involving heroin and fentanyl. Prescription opioid overdose death rates thereafter did not drop, and the drivers of overdose deaths built on one another. Opioid overdose death rates at first were driven by prescription opioids, followed by heroin and then finally fentanyl, but they built upon rather than replaced one another, and they often involved other substances.

Addiction is not a new phenomenon; rather, today’s opioid epidemic is just the latest iteration of a longer-term problem, one that requires the nation’s attention.

Recent decreases in overdose deaths are the result of a significant infusion of federal and state dollars into the nation’s treatment system and an expansion of access to medications to treat opioid use disorder. However, the disease of addiction has been so stigmatized and treated as a criminal justice and moral issue for so long, building a national treatment system will not happen overnight. The treatment system is fractured and workforce shortages are rampant. Too few medical students receive training in addiction, too few physicians practice addiction medicine, and the system that does exist falls outside the nation’s healthcare system.

Prevention efforts are also inadequate. Addiction is one of the few diseases where we expect people to diagnose themselves by “hitting bottom.” For example, despite a recommendation from the American Academy of Pediatrics, too few pediatricians talk to their patients about substance use, even though this could provide an opportunity for early intervention. Instead, the condition is ignored until it becomes chronic and more difficult to treat.

Improving the nation’s health care system to better respond to pain, as well as to substance use disorders, requires a long-term effort. People in pain, along with people with substance use disorders, can join together, build consensus, and make certain that both groups receive the type of care they need and deserve. The harm reduction services Dr. Rieder embraces should be available throughout the nation’s healthcare system as part of drug user health. Addiction can be prevented, and developing a substance use disorder, like other chronic conditions, is not inevitable. Early intervention, improving the social determinants of health, and responding humanely to people with pain and substance use disorders are all necessary to improve long-term health outcomes.


[i] See a recently released report from Columbia Legal Services in Washington state, “Gone but Not Forgotten The Untold Stories of Jail Deaths in Washington,” detailing the consequences of drug withdrawal, including opioids) in Washington state jails.

[ii] 2017 data released from the National Center for Health Statistics indicated a leveling off in prescription opioid involved overdose deaths and preliminary data for 2018 indicates a 3.4% decrease in drug overdose deaths.

The Conversation

Dr. D’Onofrio Is Correct about Stigma but Seems Too Reticent Regarding Policy Proposals

In her response to Travis Rieder’s lead essay, Gail D’Onofrio laments that stigma presents obstacles to patients seeking treatment for substance use disorder, and on that we agree. And not all people with substance use disorder desire treatment.

I have written here about the stigma attached to the nonmedical use of certain substances, pointing out that there are many “functioning addicts” who, if they live in the right neighborhood, have the economic means, and the right connections, are able to live productive and meaningful lives while keeping their lifestyle choices confidential. There is no doubt that there is a racial basis to the stigmatization of the nonmedical use of certain drugs over others. For a detailed portrayal of the role of racism in this issue, I recommend Johan Hari’s Chasing the Scream—The First and Last Days of the War on Drugs.

As I pointed out in a letter to the editor of the Wall Street Journal, the stigma attached to certain substance use disorders is responsible for many of the harms wrought by these disorders, by driving addicts to the dangerous black market and away from medical attention.

Unfortunately, it appears that Dr. D’Onofrio still subscribes to the standard narrative that the opioid overdose crisis is a direct result of the overprescribing of opioids to treat pain, largely due to an aggressive advertising campaign by Purdue Pharma, makers of OxyContin (released in 1996), and other drugmakers.

In my response to Travis Rieder, I stated that data from the National Survey on Drug Use and Health as well as the Centers for Disease Control and Prevention show no correlation between prescription volume and the incidence of prescription pain reliever “misuse” or use disorder. Furthermore, research published last year by the University of Pittsburgh Graduate School of Public health found that overdoses from the nonmedical use of licit and illicit drugs have been on a steady exponential increase since at least the late 1970s, and current data show no evidence of deviation from the trend line. An NBC News reporter who interviewed the lead author summarized the study by writing, “[The overdose crisis] started before the availability of synthetic opioids, and may have only a little to do with the prescribing habits of doctors or the pushy habits of drugmakers, the team at the University of Pittsburgh found.”

I am not here to defend the aggressive marketing strategies of Purdue Pharma. But it is important to note that there were several studies throughout the 1970s, 1980s, and 1990s criticizing the undertreatment of pain and arguing for liberal prescribing of opioids. In fact, a 1993 newsletter from the National Institute on Drug Abuse stated, “these drugs [opioids] are rarely abused for medical purposes…thousands of patients suffer needlessly,” and in 1992 and again in 1994 the U.S. Department of Health and Human Services issued guidelines encouraging a more liberal use of opioids to treat postoperative and cancer pain. I think it is undeniable that “opiophobia” was responsible for the undertreatment of pain prior to the mid-1990s. Unfortunately, in reaction to the overdose crisis, the pendulum has swung back in the other direction, and “opiophobia” is back with a vengeance today.

It has become fashionable to place the blame for the overdose crisis on doctors and drugmakers, but in fact, the problem has been going on for decades and has complex sociocultural and psychosocial etiologies. To be sure, many unscrupulous doctors and pharmacists teamed up with drug dealers to run “pill mills,” but they were the exception, not the rule—and they were not treating patients as much as they were supplying nonmedical users with their drug of choice. While they deserve some blame, the ultimate blame should be placed where it belongs: on drug prohibition and its tendency to corrupt health care practitioners, police, and politicians with the lure of easy money.

Further complicating the fashionable narrative, last month the Washington Post obtained access to a database maintained by the Drug Enforcement Administration. It revealed that Purdue Pharma’s OxyContin accounted for just 3 percent of the prescription opioid market from 2006 to 2012.

I also disagree with Dr. D’Onofrio regarding the risk of opioid misuse or addiction when used to treat acute or chronic pain. While she provides sources, I too provide sources and will argue that the 0.6 percent misuse rate with opioids prescribed for acute postoperative pain published in the January 2018 BMJ by Brat, et al, as well as the two Cochrane systematic reviews I referenced showing a roughly 1 percent addiction rate associated with long-term treatment of chronic non-cancer pain patients are more recent, comprehensive, and rigorous studies.

I agree with Dr. D’Onofrio that “harm reduction strategies are imperative” in addressing the overdose problem, and I appreciate her comparison of diabetic patients whose conditions are poorly controlled to patients suffering from substance use disorder. In fact, harm reduction per se is much of what we doctors “do” every day in modern, developed countries. As I point out here, when health care practitioners medicate overweight and sedentary patients with anti-hypertensives, statin drugs, or oral anti-hyperglycemic medications, they are engaging in harm reduction strategies for what in great part are lifestyle choices made by our patients.

While I don’t disagree with any of Dr. D’Onofrio’s policy proposals, I am disappointed that they are too acquiescent to the current regulatory regime, which is in desperate need of reform. For example, Dr. D’Onofrio calls for extensive training of hospital staff, including nurse practitioners, so they can get an “X Waiver” to prescribe buprenorphine, an effective form of Opioid Agonist Treatment. Better yet, why not call for removing the requirement for an “X Waiver,” as I and many harm reduction advocates have been urging? In France, roughly one-fifth of primary care providers treat their substance use disorder patients with buprenorphine under no stricter prescribing regulations than for any other controlled substance. There is presently bipartisan legislation being considered in both houses of Congress to remove the “X Waiver” requirement.

I was disappointed there was no mention of reforming the laws governing the prescription of methadone as an Opioid Agonist Treatment. Current regulations, including the requirement that patients take the methadone under the supervision of a clinic staff member, are onerous and impractical, resulting in reduced access to patients seeking help. In the U.K., Canada, and Australia, practitioners have been prescribing methadone as OAT to their patients since the 1970s, in much the same way that U.S. doctors can prescribe buprenorphine (minus the “X Waiver” requirement) for decades, and pilot programs involving primary physicians in Boston, MA have shown great promise.

And while I agree that bringing an addiction specialist onboard to manage pain in critically ill patients with lengthy hospitalizations can be of great value, not every patient in pain fits that description. It can be impractical and inconvenient for many pain patients to have “too many cooks” involved in their care. I think it would be better if physicians were better educated on the pharmacology and withdrawal management of opioids for the treatment of pain, much as they are when it comes to the medical management of other conditions.

Dr. D’Onofrio and I find much on which we agree. We might find more common ground if she was more inclined to challenge conventional views regarding the etiology and management of the overdose crisis.

Waive the Waiver, but Pay Attention to Regulations

I am in favor of removing the DATA 2000 waiver, as it adds another impressive hurtle to offering life-saving treatment. However, most physicians have little knowledge about prescribing buprenorphine or regulations associated with maintaining patients on buprenorphine. So at this time I would require that at most 2-4 hour learning modules be completed with test questions at the end similar to other CME offerings, and specific to the specialty.

Once all medical schools are required to incorporate this training in their curriculum, it will no longer be necessary.

That being said, I understand the issues that we must address to prevent a recurrence of the “pill mills” that sprung up after the oxycontin debacle. We need to think carefully what oversight and regulations are in place that will not impede patient access to care, but prevent unscrupulous prescribing.

There’s a lot of agreement around harm reduction. So why aren’t we doing more of it?

Something interesting is happening in response to America’s drug overdose crisis: various experts are coming out more and more strongly in support of harm reduction services, and yet many such services continue to be vehemently opposed by politicians and a portion of the public. I was very happy to see that in response to my essay in this issue, all three respondents supported broad harm reduction efforts. And yet we are far from meeting the demand for sterile syringe exchange and sufficient supplies of naloxone, and we have yet to open the first American safe injection site.

Why this sluggish response to expert-supported, evidence-based policy?

Earlier this summer, I wrote for USA Today that I have a hypothesis about this failure of Americans to embrace harm reduction, and I thought I would use this opportunity to expand a bit on that idea. In short, I think that citing the evidence in favor of harm reduction does not change the mind of every American because at least sometimes, their objections are moral rather than empirical. That is: not everyone objects to harm reduction services because they don’t believe such services can save lives; rather, they sometimes object because, whether they save lives or not, they think it’s wrong for the state to spend resources on harm reduction. And the reason they think it’s wrong is because harm reduction—by not focusing on eliminating drug use, but instead on reducing the harms of drug use—can feel like enabling what some see as bad behavior.

This sort of reasoning can be seen when a politician denounces opioid agonist treatment as “substituting one opioid with another,” or when a frustrated sheriff laments repeat uses of naloxone for the same person, saying, “It’s just reviving somebody who’s going to go back and get high the same day.” When someone complains that a safe-injection site will essentially be a state-sponsored “shooting gallery,” the same worry is being echoed.

As a moral philosopher, I hear these sorts of condemnations as laced with a particular account of ethics: one that focuses on individual responsibility. And it is often combined with a particular view of drug use: that it is morally bad. Combined—without any further views about addiction to complicate the picture—we can see the old, harmful, stigmatizing view of addiction as a moral failure: drug use is bad, and people are responsible for their choices. So if they choose to do drugs, they should live with the consequences, and enacting harm reduction policies would actually be quite problematic, as they would be helping these people to do the morally wrong thing.

I take this view to be tragically, catastrophically mistaken; but I also want people (including my allies in support of harm reduction) to notice how understandable the view looks if you have certain beliefs. Because if my experience is any indication, and if this is really one of the significant obstacles to embracing harm reduction, we must tackle this view head-on, rather than simply continuing to tell our opponents that harm reduction is evidence-based.

We could take issue with the high-level moral philosophy being invoked and argue that whether or not an individual is responsible for her actions is not actually all that morally important. For transparency’s sake, I’ll note that this is actually my own inclination. But I also think it’s both philosophical overkill and unlikely to be effective. Because anyone with strong intuitions about moral responsibility is unlikely to be convinced, and we can respond to the more specific challenge for drug policy with less grand claims. These more modest claims are: 1) that even if moral responsibility is a good foundation for ethics and policy, it doesn’t work well in cases of compromised integrity (like with addiction); and 2) that even if moral responsibility is a good foundation for ethics and policy and if addiction didn’t compromise integrity, healthcare isn’t (and shouldn’t be) organized around a principle of personal responsibility.

Regarding 1: addiction is a health condition characterized by injury to the will. That is, the precise effects on the brain that define addiction result in cravings and compulsive behavior, even in the face of adverse consequences. What it means to be addicted is to fail to be able to fully regulate one’s will. And when someone’s will is compromised, that changes our evaluation of their responsibility. This is true for children, those with other mental health conditions, or someone being controlled or coerced. It is a foundational view of moral philosophies that focus on personal responsibility that certain conditions must be met in order for ascriptions of responsibility to be appropriate. But addiction undermines full autonomy, and so the focus on responsibility for one’s actions when suffering from addiction is unhelpful and incomplete. We should note that this does not mean that someone who is addicted to a substance has no responsibility for their actions—having an addiction does not turn one into a robot or a zombie—but it does compromise one’s responsibility.

Further, argument 2 notes that even if someone suffering from addiction has sufficient autonomy to be fully responsible for her actions, the proper goal of healthcare is not to reward good behavior and punish bad behavior. Many health outcomes are related to personal choices to some degree. As Dr. Singer notes, overweight and sedentary patients are regularly treated with medications for conditions that are partially related to lifestyle choices. For another example, joint replacements are incredibly common in the United States, and they can be the result of activity choice, weight, or both. Even the common broken bones and trauma resulting from everyday activities like playing sports, hiking, skiing, or driving are health conditions that result from our free and informed choices.

Focusing especially on the case of injury from sports (and how I have never heard anyone suggest that athletes should be forced to live with consequences of their risky choices), it becomes clear that a lot of the force of the moral responsibility argument is being carried by a stigmatizing view of drug use. If athletics is good and drug use is bad, then we can separate out healthcare for those suffering from addiction as unjustified. But now it becomes clear that the view is not a rigorous moral philosophy, but an individual view about what behaviors are good and bad.

Do we really want healthcare policy to be based on what some group of people thinks is bad behavior?

I certainly don’t. I have moral objections to American football, but I certainly want all football players to have access to healthcare. And as a past motorcyclist, I was told many times in my life that I was a bad person for taking such risks when I had a family. Having now paid the price for my decision to ride, I see the wisdom in their comments, but am eternally grateful that I wasn’t denied healthcare because of my lifestyle.

The goal of healthcare is to promote health, not to reward or punish certain lifestyles and choices. We can promote the health of people who use drugs through harm reduction services, so let’s recognize that as consistent with the ideal of healthcare, even if some people disagree with their lifestyle.

Areas of Consensus on Opioid Treatment

After reading the essays and responses in this Cato Unbound issue, the differences of opinion seem to center mainly on the origins and cause of increasing rates of opioid involved morbidity and mortality seen in the United States in the last several years. This is an issue where debate will continue, particularly in light of the recent decision in Oklahoma against Johnson and Johnson, and the Ohio case against the pharmaceutical industry set for October.

Regardless of these differences, I believe we have consensus in the following areas. Stigma has resulted in policies detrimental to the long-term health and well-being of people with substance use disorder, and both harm reduction programs and access to medications to treat opioid use disorder should be expanded.

There are a multitude of programs included under the category of harm reduction, from overdose prevention sites (also called safe consumption facilities) to syringe services programs, to naloxone distribution. Syringe services programs are gaining more acceptance nationally, with states such as Florida legalizing syringe exchanges earlier this year. In Kentucky, identified by the Centers for Disease Control and Prevention as having several counties at elevated risk for outbreaks of blood borne diseases due to injection drug use, syringe services programs have become a central element of the state’s opioid response. Kentucky passed legislation in 2015 legalizing syringe services programs, and as of February 2019, 47 counties in Kentucky have programs.

A final area of consensus in these exchanges is support for increased access to medications for the treatment of opioid use disorder. One of the essays, by Dr. Jeffrey A. Singer, discussed removing the federal waiver requirement necessary to prescribe buprenorphine; equal emphasis should be applied to removing the significant barriers to methadone treatment. Methadone can only be dispensed in a specially licensed facility that is subject to a multitude of federal and additional state requirements. Calls for revising federal law and allowing the mainstreaming of methadone treatment into health care have not resulted in congressional action.[i] Reforming the entire system of methadone treatment for opioid use disorder may be years away. However, an important first step could be taken to expand methadone availability by lifting the ban on new methadone vans, a moratorium that has been in place since 2007.

The Drug Enforcement Administration (DEA) licenses methadone vans and a new rule must be issued by DEA to lift the moratorium. Methadone vans are often used in underserved or remote areas where opioid treatment facilities are unavailable. Lifting the moratorium and allowing new methadone vans will provide areas of the country in dire need of additional treatment access to this effective medication. A wholesale rethinking of laws around methadone treatment may be necessary, but in the meantime, this is an essential and urgent first step.


[i] Samet, J.H., Botticelli, M., & Bharel, M. (2018). Methadone in Primary Care - One Small Step for Congress, One Giant Leap for Addiction Treatment.The New England journal of medicine, 379 1, 7-8 .