THE ELUSIVE ADDICTION-TREATING PHYSICIAN

KELLIE HAD EXPERIMENTED with a variety of drugs as an adolescent but had never been hooked on anything. Ten years later, as a mother of three young kids and in a violent relationship with her boyfriend, she injected heroin to help her relax. Soon it was all she could think about. She loved her children, but heroin clouded everything. Each morning she would wake up wanting to quit, intending to quit, but within a few hours she was shooting up again. One day, Kellie’s sister, Monique, found the children hungry, dirty, and crying in a house littered with needles. Monique took the kids to her own house, telling Kellie she would give her a chance to “get straight” before notifying the state’s Department of Child Services.

Wanting help, Kellie voluntarily checked herself into a local detoxification center. During the few days of detox, she felt such overwhelming cravings that she could think of only one thing—her next hit of heroin. The moment she left detox, she drove straight to her dealer. In response, Monique contacted the Department of Child Services, who formally removed Kellie’s children and placed them in Monique’s care.

Once more, Kellie voluntarily entered detox. Once more, after detox the facility gave her a few brochures but made no effort to connect her to ongoing treatment. Therefore, Kellie took matters into her own hands and drove to the large community mental health and addiction treatment center downtown. She asked for addiction treatment, explaining that she had just completed detox but still felt cravings and was deeply depressed. Even though the facility technically allowed walk-ins, the approximately two-hour wait merely resulted in a rudimentary evaluation, after which Kellie was told she would receive a phone call to schedule an appointment. About a week later, Kellie did receive a phone call, but the receptionist explained that group counseling was full and was required prior to sessions with an individual counselor. The handful of staff psychiatrists was also not accepting new patients. Kellie hung up in tears.

A few hours later, she injected heroin on the bathroom floor. She thought about killing herself by purposely overdosing. The next few weeks were a blur. She did not want to get high; she just wanted to survive the next few hours. Seeing her misery, a friend undergoing Suboxone treatment with a local primary care physician offered one of his own strips. He said it would at least prevent the sickness of heroin withdrawals. A few days later, Kellie took the Suboxone and felt better almost immediately. She did not feel high, merely normal, a feeling she had not had in a long time. The friend warned her not to use heroin in the immediate future or else the Suboxone in her system would interact negatively with the heroin, making her sick.

The next day, Kellie bought another strip from the same friend, though the friend raised the price significantly. Remembering how the Suboxone made her feel, Kellie decided to start treatment on her own. She tried calling her friend’s Suboxone-prescribing physician to schedule an appointment. But the receptionist immediately said the physician was not accepting new patients. Kellie asked for a recommendation to another Suboxone provider in the area. The receptionist did not know of any, but she encouraged Kellie to find a list of buprenorphine prescribers on the Substance Abuse and Mental Health Services Administration’s website.

Kellie entered her zip code into the government website and found four other buprenorphine-prescribing physicians in her city. Her heart raced with the anticipation of help that was in sight. She called the first physician: not accepting new patients. She called the second physician: no longer prescribing buprenorphine. She called the third physician: his office didn’t even seem to know what buprenorphine was. Finally, with a sinking feeling, she learned that the fourth physician had retired.

Still determined to get help, Kellie expanded her online search to physicians located within a one hundred–mile radius. Fortunately, the first listing in a neighboring city resulted in an appointment for the following week. The only downside? It was over an hour away. But Kellie didn’t care. She could wait one week, so long as her friend kept selling her some of his own strips in the meantime.

ABOUT A QUARTER OF URBAN US counties and more than half of rural counties lack a single DATA-waivered physician,199 meaning a physician who is legally able to prescribe buprenorphine under the Drug Addiction Treatment Act of 2000.*

In 2019, buprenorphine prescribers are still heavily restricted. A DATA-waivered physician can prescribe buprenorphine for up to thirty patients at one time during the first year of the DATA waiver, and then for up to one hundred patients thereafter. A small minority of physicians who operate in “qualified settings” or are board certified in addiction medicine can prescribe it for up to 275 patients beginning in their third year.⁸³ It is clear that these restrictions are based not so much on the medication’s efficacy and safety potential, but rather on misunderstandings of the population that seeks buprenorphine treatment for addiction. Why else would the restrictions not apply to prescribing oxycodone or hydrocodone?

Many experts in the early 2000s viewed the medication’s FDA approval as a turning point in addiction health services. Office-based prescribing could dramatically expand treatment accessibility. It might even lead to patients feeling more comfortable discussing opioid addiction with their primary care physicians, who for the first time since the passage of the Harrison Act could prescribe something useful for opioid addiction treatment from their offices rather than just referring patients to an outside provider.

But by 2015, fewer than half of US counties had a single DATA-waivered physician.93 In 2018, more than a third of all US counties lacked a single DATA-waivered practitioner, and more than half of all rural counties lacked one.²⁰⁰ Furthermore, many practitioners with a DATA waiver rarely prescribe buprenorphine to any patients.^201,202 One study found that a third of providers with a thirty-patient limit never prescribe the medication;²⁰³ another found that almost half of DATA-waivered practitioners prescribe to fewer than five patients.²⁰¹

Hoping to help patients locate a buprenorphine prescriber, the Substance Abuse and Mental Health Services Administration created an online public registry of DATA-waivered physicians. But even when physicians are listed on the government registry, they may have already met their maximum number of legal patients. And furthermore, physicians can opt out of the registry, resulting in a public list with only about half of DATA-waivered physicians visible.²⁰² Physicians tell me they opt out of the public list because they do not want to be known as the “Suboxone doctor.” They do not want “unsavory” patients lining up outside their door, causing their more “respectable” patients to leave the practice. Some physicians prefer to prescribe buprenorphine to only a few trusted, handpicked patients. I have trouble imagining physicians in other health care fields hiding their practice in this way.

Like methadone, buprenorphine can technically be prescribed for pain at a different dosage and frequency, with patient limits and education requirements not applying to pain prescriptions, resulting in two parallel regulatory structures: one for stigmatized addiction patients and one more sympathetic to pain management patients. Many patients whom I have interviewed view buprenorphine patient limits as hypocritical and unfair. Ian said, “It’s infuriating . . . they have [patient limits for buprenorphine], but not Opana or Oxycontin? That doesn’t make any sense . . . why are you going to regulate the treatment and then not regulate the Oxycontin? . . . I know it gets in the way of people getting better because I [work at a syringe exchange] and probably once a week I get a call from someone saying, ‘I’m looking for a Suboxone doctor,’ and it’s a whole ordeal because you have to find someone who has a space open and who takes Medicaid or a sliding scale.”

Taking into account current medication-assisted treatment laws and prescriber capacity, approximately three times as many patients could receive buprenorphine treatment as are currently receiving methadone treatment, making buprenorphine treatment expansion a focus of some scholars and policy makers.202 Nevertheless, buprenorphine treatment capacity, meaning the number of DATA-waivered physicians who can treat patients with buprenorphine, is far too low. In 2015, almost every state (96 percent), including the District of Columbia, had an opioid misuse rate higher than its buprenorphine treatment capacity.²⁰² Why is capacity so low? Stated differently, why do so few physicians have a DATA waiver? And of those who do, why do so few prescribe buprenorphine?

It is worth noting that buprenorphine is not the only difficult-to-access form of office-based MAT. Relatively few people in recovery from opioid addiction have tried extended-release naltrexone even though it does not activate the brain’s opioid receptors, has relatively lower stigma than methadone and buprenorphine, and is subject to minimal regulations, with no patient limits or required special education for physicians. Ironically, the inability to divert the medication and sell it on the street has likely limited knowledge of extended-release naltrexone, since many in recovery report learning of buprenorphine by seeing it bought and sold illictly.178 But among those patients who want to try extended-release naltrexone, the experience of finding a provider for that medication can be just as grueling and disheartening as finding a buprenorphine treatment provider.

OPIOID ADDICTION IS A relapsing health condition not dissimilar in treatment success rates from hypertension, asthma, and diabetes.²⁰⁴ Yet people are far more likely to visit their physicians if they have hypertension, asthma, or diabetes than if they have an addiction. Instead, they think of physicians as an addiction treatment provider of last resort or are unsure what help a physician could provide at all. Self-help groups, such as Alcoholics Anonymous or Narcotics Anonymous, are by far the most common “treatment” setting, even though support groups do not say they provide treatment.⁹⁸ Never have I heard an interviewee say, “When I realized I needed help with my addiction, I contacted my primary care physician.”

Consider how unusual it would be for a patient with any other deadly, chronic medical condition to first visit peer-led, nonprofessional support groups, then rehab centers (often with few, if any, physicians), and then counselors—all before finally contacting a physician. Furthermore, imagine that once the patient finally decided to contact a physician, the patient discovered that his primary care physician, like other physicians in the area, does not treat the condition with the latest evidence-based treatments. Instead, the physician merely refers the patient back to support groups, rehab centers, and counselors.

During the course of my research, I have conducted in-depth, semistructured interviews with dozens of physicians who provide addiction treatment. In one of my studies, I explicitly asked physicians to compare barriers to prescribing the oral versions of buprenorphine, such as Suboxone, and extended-release naltrexone (i.e., Vivitrol).¹⁷⁸ But I quickly realized I was asking the wrong question. Barriers do not just exist to prescribing MAT; barriers exist to treating addiction, period. As a result, many physicians don’t treat addiction at all, regardless of medication options. So, I changed my research question to: Why do so few office-based physicians, meaning those outside of methadone clinics or hospitals, treat addiction using any method?

It quickly became clear that physicians lack sufficient training in and education about addiction medicine. As a result, they fail to see themselves as addiction treatment providers. Instead, they feel that addiction treatment is the exclusive realm of counselors, support groups, and rehabilitation centers. According to one recent study of medical students’ attitudes concerning addiction patients, the researchers found that “[the medical students] rarely mentioned addiction treatment, and when they did, they perceived it as something that was done by specialized counselors out in the community. At most, they perceived that the physicians’ role was to identify and refer.”205 The problem is not that physicians collaborate with other professionals on addiction treatment teams. In fact, the larger and more interdisciplinary the team, the better! The problem is that physicians purposely exclude themselves from the interdisciplinary team or, at most, take a minor role.²⁰⁶ But only physicians, nurse practitioners, and physician assistants can prescribe buprenorphine or extended-release naltrexone. And many states require that DATA-waivered nurse practitioners and physician assistants prescribing buprenorphine be supervised by a DATA-waivered physician.²⁰⁷

Medical schools and residency programs notoriously provide inadequate training in addiction medicine. In the words of Dr. Richardson, an internal medicine resident who only recently graduated from a major public university’s medical school, “It’s very minimal. I don’t remember any formal lecture about addiction medicine at all.” These feelings are pervasive and reported by multiple studies of medical school students.²⁰⁸ One recent survey of internal medicine residents found that 37 percent reported having received no medical school addiction instruction; of the 63 percent who had some addiction training, only 47 percent received more than a single lecture, and only 25 percent interacted with an addiction patient.209 Failure to provide more robust addiction training is inexcusable, particularly in the face of the opioid overdose epidemic. Some could argue that it amounts to educational malpractice.

Even when medical school curricula incorporate addiction medicine, it is rarely taught as an entire course,²¹⁰ instead comprising a much smaller niche, often a single lecture—grossly inadequate in light of the number of patients with addiction whom physicians will eventually see. Such meager education offerings send a loud message to medical students that addiction medicine is beneath them, unimportant, or irrelevant, despite the fact that 10 percent of their patients will experience a substance use disorder at some point in their lives.²¹¹ In one interview, Dr. Johnson complained about the many hours her medical studies devoted to rare genetic diseases versus the single lecture on addiction, despite the fact that patients with rare genetic diseases are exactly that—rare. But rare genetic disorders hold educators’ interest more than addiction science, which seems more pedestrian. She felt utterly unprepared for the many patients she would eventually see with opioid addiction.

Another internal medical resident, Dr. Palmer, had only been exposed to addiction medicine during a family medicine rotation through informal conversations with a family medicine physician who happened to have a board certification in addiction medicine. Addiction medicine was not even taught during Dr. Palmer’s psychiatry rotation: “I guess [the psychiatry rotation] is tailored toward everything except addiction medicine because that’s not on the shelf exam at the end of each rotation . . . and the stuff you learn is just sort of in passing, if you are having a conversation with one of your attendings.”

Nor are these educational deficits remedied during residency. One recent study found that 25 percent of internal medical residents felt unprepared to diagnose addiction, and 62 percent felt unprepared to treat addiction; no resident reported feeling “very prepared.”209 In another study of over one hundred medical residents, none answered all questions about addiction medicine science and treatments correctly, almost half couldn’t identify how buprenorphine treatment works on the brain, and 81 percent failed a naltrexone question; only 6 percent of residents correctly answered all MAT questions.²¹² In light of these educational deficits, it is unsurprising that physicians report feeling less comfortable prescribing medications for addiction than prescribing medications for other chronic conditions.²¹³

Dr. O’Brien, director of a large medical school’s psychiatry residency program, disclosed that until a few years ago, the school’s training in addiction medicine was limited to observations of twelve-step groups and a brief lecture. He strongly disagreed with this educational plan and advocated incorporating education about MAT, modern counseling methods, and other evidence-based treatment approaches. The prior minimal education implied that addiction treatment is not a matter for physicians but something left to support groups or counselors only.

Moreover, medical residents’ exposure to addiction treatment typically occurs in short-term hospital or acute care settings rather than in other more representative settings, such as outpatient clinics. The fact that residents usually see addiction treatment in brief episodes is particularly unfortunate: they cannot witness how addiction treatment works over time, even though addiction is a chronic, relapsing disease. Without this long-term view, residents may assume that addiction treatment does not work. Nor do residents in short-term treatment settings learn how to adjust addiction treatments in response to patient progress or setbacks, or how to transition patients to different medications.

Inadequate addiction medicine education in medical schools and residency programs reflects the fact that very few medical school faculty members are trained in addiction medicine. Yet, both curriculum instruction and education through attending physicians contribute to a physician’s perception of how effectively she can treat addiction. Moreover, addiction medicine faculty can serve as role models for students and residents, teaching them how—and how not—to interact with patients with addiction symptoms, lessons that few are currently learning.

We know that addiction medicine education works—specifically, that it increases physicians’ preparedness to diagnose and manage addiction.208 The best curricula will incorporate several features, including student exposure to patients in long-term treatment and recovery programs, student engagement in clinical work under the supervision of physicians experienced in treating addiction, and information on evidence-based treatment methods. Given the strong association between addiction, chronic pain,²¹⁴ and mental health disorders, addiction medicine education should explore the treatment intersection between these topics. In fact, about half of people who experience a substance use disorder will experience a mental health disorder during their lifetime, and vice versa.²¹⁵ Online training can also fill curriculum gaps, including programs that walk students through case vignettes in the style of “choose your own adventure.” Such training methods can dramatically increase students’ perceptions of addiction treatment effectiveness.²¹⁶

Interestingly, the eight-hour course required for physicians to receive a DATA waiver, and thus to prescribe buprenorphine, can heighten their confidence in treating addiction and their willingness to do so.²¹² Therefore, one potential route to expanding buprenorphine prescribing is to require this course, which is available online, in most residency programs regardless of whether a physician has a preexisting desire to treat addiction. For example, Dr. Capone’s psychiatry fellowship required fellows to obtain DATA waivers for prescribing buprenorphine. Although an unusual requirement, she believes it should be universal. She stated, “At the end of the day, it’s going to be a [general practitioner], it’s going be a family practice, it’s going to be the emergency room that sees the majority of these patients. So, I would like those three, in addition to psychiatry, to have [the buprenorphine certification course] as part of their training being mandatory. And in my ideal world they would all get buprenorphine waivers, so the burden doesn’t fall on one person.” This last statement reflects the fact that Dr. Capone is the only physician prescribing buprenorphine in her large treatment facility, which has hundreds of patients with opioid addiction. Eight hours, though, is arguably not enough time to address an immensely complex topic. It barely scratches the surface, but at least it is a start.

Given the United States’ current and future public health needs, the woeful state of addiction education in medical schools and residency programs is indefensible. But there are few direct incentives to remedy these shortcomings. Medical schools must remain solvent, and relative to other fields of medicine, addiction medicine is not a profitable field. Addiction medicine is low-tech and thus less “sexy” than surgery or other fields that receive constant influxes of private donations and grant money. For medical schools to expand curricula and hire faculty experts, federal and state governments must increase targeted funding for addiction medicine education, and private donors must be persuaded of addiction medicine education’s value—a tall order given that medical schools themselves often discount its importance.

But there is another way to motivate professional interest in addiction medicine: by increasing the number of addiction medicine questions on state medical licensing exams. To a large extent these tests dictate what is taught and what is studied. While “teaching to the test” is rarely the best educational strategy, tests should reflect the subjects on which professionals must be competent to practice, and addiction medicine surely meets that standard. Recent medical school graduates tell me that addiction medicine would receive greater educational attention if state medical licensing boards included the subject matter on exams. Medical schools care deeply about what percentage of students pass state licensing exams because the passage rates affect school rankings. But for addiction medicine to receive higher priority on these tests, members of state licensing boards must accord it higher significance, presenting the same dilemma that makes medical school educational reform so difficult.

Reforming medical school and residency programs, though, is not enough to remedy physicians’ lack of knowledge. After all, most practicing physicians have passed those educational milestones. For physicians already working in office-based practices, mentoring is one well-regarded method for transferring knowledge and skills. The government created a Physician Clinical Support System for buprenorphine treatment, which is an educational support system geared toward non–addiction specialists who have already undergone the mandatory eight-hour training. It’s a national network of trained physician-support mentors and includes phone, email, and in-person support.217 The problem is that few of the physicians whom I have interviewed, even those already DATA-waivered, have heard of it. Mandatory continuing medical education on the topic, at least for physicians practicing in certain fields, could help. Another option is “academic detailing,” basically a practice wherein health care practitioners trained by academics visit office-based physicians and promote certain underused treatments, much as a pharmaceutical representative would do, though with far less bias and no financial conflict of interest. While understudied in the context of addiction health services, existing research shows that academic detailing has promising results in increasing health care practitioner utilization of evidence-based treatment methods.²¹⁸ Similarly, the existence of state clinical guidelines for providing buprenorphine treatment is associated with more DATA-waivered physicians,²¹⁹ possibly because such guidelines increase physician education and confidence in addiction treatment.

Arguably all practicing physicians should be targeted with education efforts, but two groups stand out: primary care physicians and psychiatrists. The former is the most accessible group of physicians in the United States, and the latter is particularly well suited to address complex co-occurring mental health issues. Disconcertingly, some physicians whom I have interviewed seem to think that only psychiatrists should treat addiction. Yet, research indicates that people without severe co-occurring psychiatric conditions, such as post-traumatic stress disorder, or polysubstance use, such as both opioid use disorder and cocaine use disorder, do well in addiction treatment with a primary care physician or internist.220 Though such patients may greatly benefit from meeting with a psychiatrist, all patients do not necessarily need addiction treatment from one—an important point given the paucity of psychiatrists in the United States and the existing stigma related to seeing a psychiatrist. An individual obtaining treatment from a primary care physician could be, for all anyone knows, getting a flu shot. Ideally, seeing a psychiatrist would not be stigmatizing. But that is not the society we live in today.

On the other hand, people with serious co-occurring mental health conditions or significant polysubstance use are less likely to fare well outside of specialized addiction treatment settings.²²⁰ One approach is for primary care physicians to diagnose the severity of the opioid addiction and then decide whether the patient needs specialized care. If yes, the primary care physician would send the patient to the specialist. If no, then the primary care physician would proceed to treat the patient. Even if the primary care physician sends the patient to a specialist, the specialist and primary care physician should continue to share patient information and work together as a treatment team. There are many conditions related to drug use, such as HIV/AIDS or hepatitis C, for which the primary care physician can continue to provide treatment. Furthermore, once the patient has stabilized and made significant progress in the specialist setting, the treatment can be “stepped down” and the patient can return to the primary care physician’s office for maintenance medication.

Regrettably, provider collaboration in addiction treatment is impeded by their misunderstanding of privacy laws.221 In addiction treatment, two privacy laws reign supreme: the Health Insurance Portability and Accountability Act (HIPAA) and Title 42 of the Federal Code of Regulations Part 2 (42 CFR 2). HIPAA and 42 CFR 2 were created at different times for different purposes.

HIPAA, and its associated regulations, applies to all private health information and protects patient privacy but also ensures flexibility in treatment and health insurance services, as the word “portability” in the name implies. In contrast, 42 CFR 2 was written decades before HIPAA with the purpose of preventing potentially stigmatizing substance use and mental health information from reaching unintended eyes and ears. It does not apply to health information generally, singling out and adding an extra layer of protection to substance use information specifically.

As a result, HIPAA and 42 CFR 2 differ in the extent to which they allow health care providers to share information about patients for purposes of treatment collaboration and integration before the health care provider must get special consent from the patient. Special consent means extra paperwork and associated liability risks. For example, HIPAA allows health professionals to share patients’ private health information without obtaining patients’ consent when the purpose of sharing information is for referral, consultation, or coordination of services.^222,223 This rule applies to all health conditions except substance use disorders.

For substance use disorders, providers must follow privacy regulations in 42 CFR 2. As originally written, 42 CFR 2 required treatment providers to obtain specific written consent from patients prior to sharing health information with other treatment providers, only allowing for a few narrow exceptions, such as life-threatening emergencies. The patient had to complete a document specifically naming the health care practitioner who would receive the information, defining the type of information to be shared, and stating the dates during which information could be shared. Each time a new health care provider came into the picture, such as a new counselor or primary care physician, a new form would need to be created and signed by the patient.

One might wonder why consent forms would serve as any kind of barrier to begin with. Isn’t it easy for a patient to complete and sign a form? The problem is not so much on the patient’s side but more so on the provider’s side, as each additional layer of paperwork is time-consuming, requiring staff resources. Furthermore, 42 CFR 2 applies to substance use–related health data only, rather than to other health conditions, making the former appear more legally risky to collect and share. As my interviews with health care providers reveal, physicians and counselors are very sensitive to potential liability risks, even if these risks are theoretically easy to overcome or very remote. Additionally, most electronic health record systems do not allow for data segmentation, meaning separation of substance use–related health data from other types of medical data. Data segmentation would allow providers to share some parts of the electronic health record while hiding other parts, such as substance use data. But without data segmentation in the electronic health record systems, providers may have difficulty sharing any health information so long as substance use–related health data is embedded within the medical record.221

In response to numerous complaints about the tedious consent process involved in treating substance use disorders, in 2017 the US Department of Health and Human Services amended 42 CFR 2 to allow patients to sign a general consent form in which patients can essentially say to their provider, “I allow you to share my health information with any current or future treatment provider for treatment purposes,” thereby eliminating the need for new consent forms each time a new provider becomes involved.²²⁴ But existing evidence suggests that this change is largely unknown or misunderstood among treatment providers, who still think they need new consent forms for each new provider.²²¹

Dr. Dennis Watson at the University of Illinois at Chicago has seen firsthand how both treatment services and research can be derailed by misunderstanding 42 CFR 2. In one of his projects, which involves connecting people in the emergency department who recently overdosed with peer support specialists, different floors of the same hospital are sometimes reluctant to share patient information. He says the health care providers often assume there are legal problems with information sharing even when there are none. The barriers can be even more extreme when communication involves parties from outside institutions, such as community health clinics trying to obtain data about hospital patients. Ultimately, it is the patient who suffers from this lack of coordination and integration.

Studies have demonstrated the feasibility and efficacy of starting recently overdosed patients in the emergency department on buprenorphine,225,²²⁶ after which the patients are immediately transferred to a low-barrier buprenorphine treatment provider. Such short-term (up to seventy-two hours) buprenorphine provision without a DATA waiver is legally permissible.²²⁷ But the buprenorphine treatment will automatically end in the hospital if the hospital and local treatment providers are hesitant to share patients’ private health information, missing a potentially huge opportunity to prevent overdose reoccurrence.

Surprisingly little is known about whether the extra layer of privacy afforded by 42 CFR 2 actually encourages patients to seek treatment. Based on conversations with numerous people in recovery who have never even heard of HIPAA, I find it even less likely that most people in recovery have heard of 42 CFR 2. That begs the question why the regulation exists to begin with. If its purpose is to make people feel comfortable seeking treatment but most people are unaware of it, then are the regulations helping people seek treatment or are they merely deterring provider collaboration and integration? As one scholar has written, “The [42 CFR 2] regulations predate the creation of electronic health records and reflect a time when individuals with alcohol and drug use disorders were treated almost exclusively in stand-alone, specialized facilities.”221 In a different era when federal and state governments are pushing for greater integration of care, maybe it is time to heavily revise 42 CFR 2 or even eliminate it entirely, relying on HIPAA instead.

STIGMA TOWARD PEOPLE with addiction is pervasive, not just among the general public, but even among health care professionals. It adds yet another layer of barriers to the mountain of barriers to accessing addiction treatment. Although the American Medical Association has called addiction a disease for decades, some physicians are biased against people with addiction, mirroring the American public’s biases. According to one survey, medical students frequently viewed addiction as a personal choice evolving from enjoyment, ignorance, or apathy; these misconceptions prompted medical school students to feel anger toward those who continue to harm themselves with drugs.²⁰⁵ As one medical student wrote, “I think it will be difficult to deal with people who don’t want to take an active role in improving their health.” These misconceptions at best translate into poor physician-patient interactions, and at worst they will produce stunted professionals. According to the study, “[Medical students] did not appear to react as physicians when confronted with addiction patients. They did not discuss clinical features, diagnosis, medical treatment, or advice. . . . They did not convey a sense of therapeutic optimism and pride in their acquisition of clinical skills needed to help addicted patients.”²⁰⁵

Such biases persist among practicing physicians. Addiction’s nature as a relapsing condition can be especially frustrating for physicians who want to manage it but lack the appropriate tools or training. Other chronic relapsing conditions that physicians treat routinely, such as hypertension, carry less stigma, and physicians receive comprehensive training in their management. As Dr. Andrews, a primary care physician, explained, it becomes easy for the physician to give up, thinking, “Well, those people have weak personalities, or their upbringing wasn’t good.” Treatment failures are easier to excuse if addiction is framed as a personality issue and not a medical condition.

Moreover, physicians may also worry that patients with addiction will monopolize appointments, changing clinic environments and driving other patients to new providers.178 According to one psychiatrist, “I think the fear is, if I open the door a crack, it’s going to be overwhelming. My office staff is going to be inundated with all these addicts with their needles and their waiting. This, I think, is a fanciful fear. What I want to tell them is a good twenty percent, thirty percent, even forty percent of your patients are already struggling with addiction and have not told you.” Indeed, a recent study found that the main reason physicians do not seek a DATA waiver is “not wanting to be inundated with requests for buprenorphine.”²²⁸ Physicians may also worry that the stigma surrounding addiction will rub off on themselves. As Dr. Cohen explained, “Stigma rubs off and so if you’re a patient with a psychiatry disorder there is still a stigma associated with that. There’s even more with people who are addicted, if you can believe it. The people who treat them are seen as kind of the dregs of medical practice.”

Relatedly, much research suggests physicians rarely want to treat patients with addiction because they are supposedly “difficult” and act in undesirable and antisocial ways, such as being manipulative and dishonest.^178,229 My research suggests that the fear of “difficult” patients may be related to lack of training. Unlike physicians with little experience treating addiction, physicians trained in addiction medicine are often quick to point out that “difficult” patients become great patients once the disease is managed. In fact, treating addiction often turns into one of the most rewarding experiences of the physician’s career. The physician witnesses the patient’s entire life turned around: from finding a job, to rekindling relationships with significant others and family members, to leaving the revolving door of the criminal justice system.

Physicians must learn that difficult behaviors associated with addiction are really symptoms of the medical condition, but this learning process requires empathy for patients, meaning a willingness to put oneself in patients’ shoes and to imagine how patients experience the world. Empathy is critical to treatment relationships and enhances treatment success. Contact and interaction with a stigmatized population increases empathy for the population; so, increasing medical students’ and residents’ interactions with patients who have addiction may increase empathy for this vulnerable group. Professional empathy training also may help physicians understand why patients with addiction act out, learn not to take such actions personally, and acquire skills to manage them. Though understudied, empathy training may take place in small groups of physicians who debrief after patient meetings to discuss what happened and why. Improved communication skills training could also improve physicians’ interactions with patients, easing potential confrontation and anger toward patients. Patients confronted with an annoyed or hostile physician may feel mistreated and stigmatized, pulling back from the relationship, possibly never seeking addiction treatment again.

Even physicians who want to treat patients with addiction may find that their institutions or coworkers would prefer to refer these patients out. In one study, as many as 42 percent of primary care physicians cited resistance from practice partners and 36 percent named institutional resistance as barriers to buprenorphine treatment.230 Such resistance may stem from common misconceptions about addiction and MAT. For example, if institutions or coworkers view buprenorphine as “just another drug,” then physicians prescribing buprenorphine are seen as “enablers.”

Such myths are reinforced by the perception that buprenorphine is widely misused or abused. In fact, some people do purchase buprenorphine illicitly on the street or over the internet, but they typically do so to prevent withdrawal symptoms.^61,72,231–²³³ The government reports that when people with opioid addiction experience opioid withdrawal symptoms, the symptoms are typically so strong that they return to using opioids.234 For heroin users, withdrawal symptoms may appear as soon as four hours since their last heroin use.¹⁸⁰ Buprenorphine can help fill that gap. For example, when I ask Stan, who has a history of illicit buprenorphine purchases, why he and others have bought buprenorphine on the street, he responds, “It’s definitely about preventing withdrawal. It’s usually people who are like, oh, I’m going to get clean. And so they put aside ten dollars each day or whatever it is, and they find themselves, seven days’ worth of strips. A lot of times these people fail because they don’t have the surrounding support. But I’ve had friends who have done it.”

Street buprenorphine purchases can also bridge people into formal treatment, assuming they can find a provider and pay. Some people do purchase buprenorphine illicitly to get high but that is exaggerated, especially by criminal justice administrators. As Stan further explained about buprenorphine, “It’s a miserable [high]! There’s a joke kind of street name called subaaaaaaaaahxone—like a puking noise—because if you take it and you don’t need it, it’ll make you really ill and it will make you throw up. It’s a really unpleasant high. Like if you take enough of it for yourself to feel something, you’re usually also going to feel really nauseous . . . get headaches . . . feel unpleasantly warm.” Nevertheless, misunderstandings about illicit buprenorphine use may prompt providers or their institutions to avoid prescribing it, especially if they fear that it may cause a liability risk.

Some institutional characteristics significantly influence whether a clinic is likely to adopt MAT. For example, clinics that are more likely to adopt MAT are accredited, are hospital-affiliated, provide inpatient care, and/or provide detoxification services.^235,236 In contrast, institutions promoting abstinence-only treatment are particularly resistant to adopting MAT,¹⁵⁰ as are institutions that lack affiliations with hospitals or other medical practices, perhaps due to lower physician involvement in addiction treatment and separation from mainstream medicine in general.

Changing attitudes within a health care institution takes time. For institutional changes to trickle down from the top levels of health care bureaucracies, employees in lower levels of bureaucracies must be exposed to the new practice, perceive a need to change existing practices, decide to adopt the new practice because it is demonstrably useful, and begin using the new practice. This process may take years from start to finish.

Sometimes institutional change begins not at the top but at the bottom, from providers. Dr. Miller works for a large addiction and mental health treatment organization serving hundreds of patients with opioid addiction. Just three years ago, the organization was a classic abstinence-only institution that explicitly barred clients undergoing MAT from participating in group counseling or support groups on the premises. As a former employee of both a methadone clinic and an office-based buprenorphine treatment practice, Dr. Miller brought extensive addiction treatment experience with her into her new work environment. Despite having no formal leadership role, within a few years Dr. Miller had dramatically changed institutional beliefs about MAT, partially through organizing educational MAT seminars for all staff members, including counselors, physicians, and administrators, and partially from one-on-one conversations with resistant administrators. It was not long before she received administrative permission to provide MAT, becoming the first physician in the practice to prescribe buprenorphine and extended-release naltrexone.

According to Dr. Miller, colleagues’ and administrators’ acceptance of MAT grew after they saw positive treatment results. Yet a lingering resistance to MAT still exists, particularly from peer support specialists, who are often in recovery themselves through abstinence-only methods. Sometimes they misinform patients, telling them to wean off MAT as quickly as possible, contrary to best medical practices. Thus, Dr. Miller must continually remind colleagues that long-term use of MAT is more effective than short-term use. In her ongoing attempt to change institutional culture and practices, she stresses the importance of providing regular, ongoing education about MAT to all staff members. High workforce turnover rates also necessitate such repeated educational efforts.

Finally, institutional cultural change can be coerced. Large public treatment centers frequently receive at least some, if not most, of their funding from federal, state, or local governments. This gives the government carrots and sticks to persuade treatment centers to adopt evidence-based policies. Not surprisingly, one national study found that when state mental health authorities made public funding of addiction treatment centers contingent on allowing MAT, MAT implementation increased.237

SUPPOSE THAT A PHYSICIAN has a DATA waiver and works in an institution supportive of addiction treatment, including MAT. Further suppose that this physician has adequate addiction medicine education and empathy for patients with addiction. Surely this physician will be willing to prescribe medications for treating addiction, such as buprenorphine or extended-release naltrexone, right? Unfortunately, other layers of barriers may still remain. One of the most common is insurance restrictions.

Before 2008, insurance companies often made it harder for patients to access treatment for mental health and addiction than treatment for other chronic health conditions. Such disparities prompted Congress to pass the Mental Health Parity and Addiction Equity Act of 2008, which prohibits differences in treatment limits, cost sharing, and in-network/out-of-network coverage between treatment for physical illness and treatment for mental health or addiction. A few years later, the Affordable Care Act broadened the Mental Health Parity and Addiction Equity Act so that it now applies to Medicaid-managed care organizations, the Children’s Health Insurance Program, small and large employer-funded plans, commercial health insurance sold on the Affordable Care Act marketplace, and Medicaid Alternative Benefit Plans.238 Furthermore, the Affordable Care Act required most health insurance plans to cover treatment for mental health disorders and addiction as an essential health benefit.²³⁸

But it is increasingly clear that the Mental Health Parity and Addiction Equity Act has not truly created parity for mental health and addiction treatment. Many health insurers continue to place onerous burdens on people seeking treatment for these conditions. Burdens on patients translate into burdens on physicians who try to help patients get treatment covered. The physicians I have interviewed overwhelmingly agree that parity does not exist. Some responses have been tinged with anger, such as Dr. Curry’s response: “Oh, is there supposed to be parity? I’ve never seen it!” At addiction-themed conferences, providers openly air their grievances about insurance burdens, including how insurance companies waste their time and staff time with unnecessary paperwork, preventing them from caring for patients. The highest level of anger is directed toward insurance prior authorization requirements.

Completing a prior authorization form may sound simple, but it is actually very time-consuming, with separate forms required for each patient on MAT. One study of DATA-waivered physicians who accept Medicaid found that prior authorizations were the highest rated barrier to prescribing buprenorphine.²³⁹ Furthermore, prior authorization requirements for buprenorphine may increase relapse rates, at least in the short-term.²⁴⁰ Dr. Lopez described how his office had to hire an additional billing specialist just to complete preauthorizations and argue with insurance companies that denied coverage. Other physicians decided to stop prescribing altogether. Unfortunately, prior authorization requirements are widespread. A 2015 study found that 94 percent of state Medicaid programs require prior authorization for MAT,²⁴¹ and a 2016 study found that 85 percent of private insurance plans nationwide require prior authorization for extended-release naltrexone.²⁴² But recently some states, such as California²⁴³ and Pennsylvania,²⁴⁴ have begun removing prior authorization requirements from their Medicaid programs in response to the escalating opioid overdose crisis.

If seeking prior authorizations weren’t bad enough, additional insurance rules compound this burden for physicians. Since addiction is a chronic medical condition, physicians must regularly renew each patient’s MAT prior authorizations—and insurers don’t necessarily accept renewals until the previous prior authorization has expired, preventing physicians from submitting prior authorization forms ahead of time. These delays in accessing treatment can be deadly. If a prior authorization expires on January 1 and a patient runs out of buprenorphine that day, she might not be able to receive a prescription by January 3 unless the prior authorization is submitted immediately. Buprenorphine only remains in a patient’s system for two to three days before the patient begins undergoing withdrawal and cravings return, putting the patient at serious risk of relapse and overdose.

Additionally, when completing prior authorization forms for buprenorphine, physicians are sometimes required to tell an insurance company how soon they plan to wean the patient off MAT. Some insurers require physicians to promise in writing to wean the patient off after only six months or one year. But medical studies conclude that long-term, not short-term, buprenorphine treatment is more effective.240,²⁴⁵ Arbitrary treatment termination deadlines also increase relapse rates.²⁴⁵ In response, some physicians tell insurance companies they will wean patients off MAT without actually intending to do so, or they attach a note describing their displeasure with the insurance requirement. After the six-month or one-year period has elapsed, the physicians must once again participate in the prior authorization process. According to some physicians I have interviewed, insurers rarely require provider promises to wean patients off of other medications, including prescription painkillers such as oxycodone. Therefore, these “weaning” requirements appear to violate federal parity law.

Insurance providers also have other arbitrary rules about what they will and will not pay for. Multiple physicians have reported that their state Medicaid program won’t pay for extended-release naltrexone stored in physicians’ offices. Instead patients must first attend a physician appointment to receive a diagnosis and a written prescription for extended-release naltrexone; next, the patient must go to a pharmacy to pick up the medication or have it mailed from a specialty pharmacy, which can take days; finally, the patient must return once more to the physician’s office to have the medication administered for the first time, called “induction.”

For extended-release naltrexone patients, induction requires that the health care provider inject the medication into the patient, potentially followed by a two- to four-hour observation period. Unfortunately, patients who are required to leave the physician’s office to visit a pharmacy or wait for a mail delivery before induction may never make it to the pharmacy or back to the physician. Even a simple trigger, such as seeing a building where the patient has previously gotten high, may overwhelm his or her motivation. Addiction often gives patients only a short and critical window of time during which they are willing and able to undergo treatment. If that window disappears, it may be a long time until a new window appears, and in the meantime, the patient risks overdosing.

As if arbitrary and unfair insurance rules weren’t enough, low addiction treatment reimbursement rates may further deter physicians, especially within Medicaid programs.228 Medicaid covers most addiction treatment in the United States, since addiction is associated with lower incomes.²⁴⁶ But Medicaid reimbursement rates for physicians are significantly lower than commercial health insurance rates and Medicare rates.²⁴⁷ Even commercial health insurers often reimburse addiction treatment at lower rates than other covered procedures.²⁴⁸ Addiction treatment appointments do not involve expensive surgical procedures or machines; instead, they rely on physician-patient counseling, monitoring, and medication prescribing. None of these procedures garner high reimbursement rates to begin with, a problem that plagues behavioral health care in general.

Low reimbursement rates for addiction treatment, especially in Medicaid programs, are not inevitable. They reflect a government policy choice and the stigmatization of behavioral health care and low-income people in general. In France, in contrast, physicians are paid more by the government for management of chronic diseases than treatment of other diseases, and addiction is considered a chronic disease.64 Higher physician reimbursement for addiction treatment helps counteract existing stigma toward patients with addiction.

Many Americans fall in an insurance gap in which they are ineligible for Medicaid because they make too much money for the program that targets the truly indigent, but they lack commercial insurance through an employer and cannot afford insurance on the individual market, which can be very expensive. Classic examples of people who fall in this gap include part-time employees, such as restaurant servers, or individuals who work full-time for small businesses that cannot afford to provide health insurance to their employees. For these Americans to afford addiction treatment and for physicians and treatment centers to accept them, government block grants are critical.²⁴⁹ Such grants are particularly important in states that refused to expand Medicaid, thereby often leaving single, childless men ineligible for the insurance. Grants enable health care institutions to accept patients without health insurance on a sliding scale of payments depending on their income. One study found that such block grant funding for MAT was associated with more DATA-waivered physicians.²¹⁹ Whether from the state or federal government, such grants offer an opportunity to promote best practices if required under the grant.

Encouraging more physicians to treat addiction will depend on how Congress and state legislatures react to insurance-imposed barriers to addiction treatment and whether they increase reimbursement rates for behavioral health care. Government agencies should regularly assess insurance barriers to mental health and addiction treatment. Insurers whose policies violate federal parity laws should face severe consequences. Currently, enforcement of federal parity laws occurs in a piecemeal, uncoordinated manner, and patients and health care providers may not know whom to turn to when facing discrimination from insurers. For example, typically state insurance commissioners enforce federal parity laws for insurance plans with less than fifty-one employees; the Department of Labor and the Internal Revenue Service share enforcement authority for plans subject to the Employee Retirement Income Security Act; and the Department of Health and Human Services enforces self-funded, nongovernmental plans, as well as some small group plans. Such patchwork enforcement should be simplified. How is the patient or health care provider to know whom to contact in the event of a parity law violation?

AS FRUSTRATING AS INSURANCE barriers are for physicians, they harm patients more directly, sometimes leading them to cease addiction treatment altogether or to engage in sketchy system work-arounds, such as illicit street purchases. When insurance coverage is inadequate, patients with limited means must decide whether to purchase medical care or food, or even whether they should sell some of their medication to afford the remainder of their treatment.

In early 2017, Brianna began work as a cashier at a gas station in a Midwestern city. It was the first stable job she had had in a long time. She is very proud of her progress, having already advanced from cashier to assistant manager when I interviewed her in the spring of 2018. She has not missed a single day of work nor has she ever arrived late. But the slight increase in her paycheck associated with her new position as assistant manager has made her ineligible for food stamps and her state’s Medicaid program. Brianna doesn’t really mind losing the food stamps because she can mostly afford her groceries these days, but the loss of Medicaid has hit her very hard. She is paying for commercial insurance, but the deductible is $3,000, a nearly impossible amount for her to afford. After starting her new insurance plan, she went to the pharmacy to pick up the buprenorphine prescription for which she had previously only paid $20 per month. She learned she must now pay $300 per month until she meets the deductible. Since she couldn’t afford the entire month’s supply, she asked for a five-day supply instead until she could figure out what to do.

Currently Brianna has a prescription for two daily strips of a buprenorphine, but she rarely takes the full daily dose. “Right now, I’m doing everything I can to just to stay on it, and there are days that I’ll do one or one and a half [strips] or just try to keep them, so that way I don’t run out, and I don’t have to go without. I will take less, so that way I can make it last a little bit longer.” Unfortunately, taking too low of a dose is risky. It can lead to withdrawal symptoms, and cravings can reemerge. Brianna explains, “There’s days that I have to take two because I don’t know if my body can take it. I don’t feel good.” She had quit the medication entirely once before, did fine for a while, but ultimately relapsed. Now she’s determined to continue the treatment no matter what. Brianna’s even considered going to the emergency room, just to rack up a high enough bill at the hospital that it would cover her deductible. Then she could get on a hospital income-based repayment plan. Unfortunately, she explains, she doesn’t have any condition that would qualify as an emergency.

LIKE BRIANNA, ANDY BEGAN Suboxone treatment for his oxycodone addiction while in the Midwest. He believes Suboxone saved his life, completely stopping his misuse of opioids. Over three years, he decreased his Suboxone dose from 16 mg daily to 2 mg daily. Though he no longer had cravings at the 2 mg dose, he decided he still wanted MAT, just in case. He thought Vivitrol would be a great option. He was tired of taking the Suboxone film every morning, and he wasn’t going to miss the nausea that sometimes came with it. He also wasn’t going to miss the strange looks he regularly received from the local CVS pharmacist, whose attitude toward him always dipped after seeing the medication name on the prescription paper. And no other CVS in the city carried Suboxone.

He decided to switch to Vivitrol as he was moving across the country for a new job. Now located in a large metropolis, he thought finding a Vivitrol provider would be easy, but in some ways it proved more difficult than starting Suboxone. First, he needed to undergo inpatient detoxification. He was terrified of undergoing detoxification outside of a treatment center, even from the low dose of Suboxone, after having read horror stories on the internet of painful experiences from people who tried to detoxify alone at home. But Andy’s commercial insurance only covered inpatient detoxification for alcohol addiction, not opioid addiction. Detoxification from alcohol is potentially deadly, since some people get life-threatening seizures, while detoxification from opioids is not itself deadly even though relapse, of course, can be. Not giving up, he tried a different tactic. Since he had noticed that he’d been drinking too much lately, he asked the insurance company to pay for alcohol detoxification with the co-occurring condition of opioid addiction. Now his insurance was willing to pay, though it left Andy with a coinsurance bill of almost $1,000, something he is still paying off.

Even though detoxification was a necessary step to beginning Vivitrol, the detoxification center did not actually offer Vivitrol. In fact, it offered nothing at all, except group counseling and on-site AA meetings while monitoring him and providing palliative care. With detoxification completed, he began the hunt for an outpatient Vivitrol provider. According to an internet search, about a dozen existed in his city, but by calling them he learned that most either only accepted cash payments or were not accepting new patients. Finally, he found an outpatient clinic that treated addiction and claimed to offer Vivitrol. What it actually offered, though, was an intake appointment, at which a counselor asked a series of questions meant to determine whether Vivitrol might be an appropriate medication. Determining this to be the case, the counselor then passed Andy off to a psychiatrist in a different office—an appointment that was not available until two months later. Andy met with the psychiatrist, who prescribed Vivitrol but still did not actually provide the medication. Instead, the psychiatrist sent Andy’s information to a special-order pharmacy, as per insurance requirements, which then mailed the medication directly to a third location—where the medication would finally be administered by nurses. But the third location, a “home infusion center” did not have any openings for another month. In total, by the time Andy received his first Vivitrol injection, over four months had passed since detoxification. Fortunately, Andy was stable and did not relapse, but he is deeply concerned about others less stable than him going through this process. He is almost certain that they would not make it.

* Note that having a DATA-waivered physician in your county does not mean that buprenorphine is actually accessible, merely that at least one physician is legally able to prescribe it. Whether he or she prescribes it, is accepting new patients, and accepts your insurance are totally different matters.