The United States still has the highest rate of incarceration of any nation in the world, despite a modest decline over the last decade.1 During that same decade, an opioid crisis was brewing, and has now captured the nation's attention. In 2016, an estimated 42,000 people died from opioid overdoses, 28 percent more than the year prior.Reference Ingraham2 This epidemic is particularly acute in American prisons and jails, where more than half the population meets the criteria for drug abuse or dependence.Reference Bronson3 Drug use in correctional institutions leads to a series of negative consequences during and after incarceration — not only deadly overdoses, but also the transmission of dangerous diseases like hepatitis C and HIV through injection drug use, and other long-term health consequences like pulmonary and heart infections.4
It is clear that opioid use disorder — “a problematic pattern of opioid use leading to clinically significant impairment or distress”5 — puts the lives and well-being of incarcerated people at risk. Yet in the face of such great need, only a small percentage of incarcerated people with opioid addictions have access to what leading medical organizations take to be the standard of care: medication-assisted treatment (MAT).Reference Williams6 MAT typically uses medications that are opioid agonists, such as methadone and buprenorphine-naloxone. Methadone and buprenorphine are long-acting medications that offer a slow release of dopamine and help patients refrain from illicit opioid use without the physical and psychological consequences of withdrawal or the euphoric effects of short-acting opioids.7 Both allow patients to pursue normal activities of daily living without debilitating drug cravings.Reference Park and Friedmann8
Despite 50 years of evidence demonstrating the effectiveness of MAT, “of the nation's 5100 jails and prisons, fewer than 30… offer opioid users the most proven method of recovery: administering methadone or buprenorphine.”Reference Friedmann9 Even in facilities that purport to offer MAT, few incarcerated persons actually receive treatment.Reference Chandler10
Things may be starting to change. In 2016 Rhode Island became the first state in the nation to implement a comprehensive MAT program for its entire prison system. A year and a half later, initial results show an almost four-fold increase in the number of patients receiving MAT and a 61 percent decrease in overdose deaths among people released from incarceration within the past year, a population particularly vulnerable to deadly overdose.Reference Greene11 Other states, including Connecticut and Vermont, are currently experimenting with programs of their own.12
This article argues that correctional institutions violate the Eighth Amendment when they refuse to establish MAT programs and prevent doctors from exercising medical judgment to properly treat incarcerated people with OUD. The scope of this argument is limited in three ways. First, by correctional institutions, we mean only those institutions that incarcerate people who have been convicted of a crime. We bracket questions about MAT for pre-trial detainees and other forms of detention, civil commitment and immigration detention. Second, we use MAT only in reference to treating only OUD, even though medication-assisted treatment exists for other addictions including alcohol and tobacco. These treatments are not the focus of our argument, but the structure of our argument could be useful in analyzing the constitutional requirements for treatment of those conditions. Third, the goal of this paper is not to provide a comprehensive analysis of legal claims about OUD and MAT,Reference Lebowitz13 but rather to develop a targeted argument that withholding MAT as an option for long-term treatment in prisons violates the Constitution.
This article argues that correctional institutions violate the Eighth Amendment when they refuse to establish MAT programs and prevent doctors from exercising medical judgment to properly treat incarcerated people with OUD.
In Part I, we explain what an individual prisoner or class of prisoners must show to obtain relief on a claim of inadequate medical care under the Eighth Amendment. The plaintiff must show that correctional officials showed “deliberate indifference” to a “serious medical need,” the meaning of which we examine. Part II establishes that OUD is a serious medical condition, for which long-term MAT is the most effective treatment. Yet correctional institutions largely choose not to provide it for reasons that have little to do with medical judgment. We conclude that refusing, on non-medical grounds, to provide MAT as an option for health care professionals to treat OUD evidences deliberate indifference to incarcerated patients' serious medical needs. Finally, to illustrate that provision of MAT in prisons and jails is both feasible and effective, Part III describes the comprehensive MAT system in Rhode Island's correctional facilities. Rhode Island's example shows that providing MAT in correctional facilities is not rocket science or a revolution in correctional health care, but a feasible step required by both common decency and the Constitution.
I. The Requirements for Medical Care under the Eighth Amendment
A. The Deliberate Indifference Standard
The Eighth Amendment to the United States Constitution prohibits the infliction of “cruel and unusual punishments.”14 The Court's jurisprudence on the issue began with the landmark case of Estelle v. Gamble, which established that the government takes on the burden of providing medical care when it chooses to incarcerate, because incarceration removes an individual's ability to access alternative care.15
In Estelle, the plaintiff was unloading a truck during a prison work assignment when a 600-pound bale of hay fell off and struck him.16 A month after the accident, though his back hurt “as much as it had the first day,” the prison doctor certified Gamble to return to “light work.”17 Prison officials later refused a doctor's order to move Gamble to a lower bunk and lost one of Gamble's prescriptions for four days.18 After refusing to work on account of his pain, Gamble was kept in solitary confinement for more than two months, all the while continuing to suffer from “severe back pain,” high blood pressure, and chest pains.19 He thereafter brought suit under 42 U.S.C § 1983 alleging a violation of the Eighth Amendment.
The Court held that prison officials violate the Eighth Amendment if they show “deliberate indifference” to a patient's serious medical needs.20 For officials' actions or inactions to constitute “deliberate indifference,” they must result in the “unnecessary and wanton infliction of pain”;21 an “inadvertent failure to provide adequate medical care” does not rise to the level of deliberate indifference.22 The care provided, or lack thereof, must be “repugnant to the conscience of mankind” in order to constitute a violation of the Eighth Amendment, meaning that medical malpractice alone is not a violation.23 A long line of cases after Estelle has clarified the meaning of “deliberate indifference” and what qualifies as a “serious medical need,” which we discuss below.
Before doing so, we note that the prison health care jurisprudence has developed from focusing mostly on individual claims, such as in Estelle, to including class claims raising systemic problems. For example, in Madrid v. Gomez, a class of incarcerated people at Pelican Bay State Prison in California sued over what the court called the prison's “distinct pattern of sub-standard care.”24 The court found that prison officials were deliberately indifferent because they had “abundant knowledge of the inadequacies” of the health care system and failed to remedy them, thereby “practically insur[ing] that inmates would endure unnecessary pain, suffering, debilitating disease, and even death.”25
In Coleman v. Wilson, another federal court held in favor of a class of mentally ill patients who challenged the mental health care system in California's prisons. The court held that incarcerated patients faced “an objectively intolerable risk of harm as a result of … gross systemic deficiencies” that were so obvious as to make officials' claims of ignorance unbelievable.26 The many suits challenging California's prison health care systems culminated in 2009 when a three-judge panel ordered California to reduce overcrowding in its prisons because it found that overcrowding was causing the constitutional violations at issue in the medical and mental health care system.27 The Supreme Court upheld the panel's decision in Brown v. Plata, revealing the power of systemic claims under the Eighth Amendment.28 In Plata, the Court made clear that a prison which “deprives prisoners of basic sustenance, including adequate medical care, is incompatible with the concept of human dignity and has no place in civilized society.”29
In these class actions raising systemic challenges, “deliberate indifference can be evidenced by repeated examples of negligent acts which disclose a pattern of conduct by the prison medical staff or it can be demonstrated by proving there are such systemic and gross deficiencies in staffing, facilities, equipment, or procedures that the inmate population is effectively denied access to adequate medical care.”30 In 2017, a district court in Alabama found the state's “horrendously inadequate” prison mental health care system violated the Eighth Amendment.31 In a previous decision denying defendant's motion for summary judgment, the court rejected the state's defense, writing that provision of “some treatment” is not necessarily provision of “constitutionally adequate treatment.”32
Class actions have also been brought when a group of individuals has been denied a particular type of treatment. Recently, a number of courts have certified classes of prisoners who have been denied the appropriate treatment for their hepatitis C,33 and at least one court has issued an injunction in favor of the class, in that case ordering the Florida state prison system to begin properly treating hepatitis C in its facilities.34
1. What Qualifies As Deliberate Indifference?
In the 40 years since Estelle, the Supreme Court has clarified that in order to show deliberate indifference, plaintiffs must prove subjective intent to inflict pain on the part of prison officials.35 “[A] prison official may be held liable under the Eighth Amendment for denying humane conditions of confinement only if he knows that inmates face a serious risk of harm and disregards that risk by failing to take reasonable measures to abate it.”36 The deliberate indifference test, then, asks whether a corrections officer or medical official has a “sufficiently culpable state of mind” such that the failure to treat the serious medical need is cruel.37
The standard for deliberate indifference is akin to criminal recklessness — when an individual “disregards a harm of which he is aware.”38 However, “a factfinder may conclude that a prison official knew of a substantial risk from the very fact that the risk was obvious.”39
Many lower courts have followed the Eleventh Circuit in defining deliberate indifference as: “(1) knowledge of a serious medical need and a failure or refusal to provide care; (2) delaying treatment for non-medical reasons; (3) grossly inadequate care; (4) a decision to take an easier but less efficacious course of treatment; or (5) medical care that is so cursory as to amount to no treatment at all.”40 A plaintiff can use circumstantial evidence to show obviousness, but prison officials can rebut that presumption with evidence that they were unaware of the danger.41
Circuits and individual courts often diverge in their interpretations of what constitutes care inadequate enough to evince deliberate indifference. For example, disagreeing with a doctor's choice to prescribe a particular medication over another is typically insufficient evidence to support a deliberate indifference suit, even if that decision fell below the “standard of care.”42 Indeed, courts diligently avoid second-guessing the professional judgment of a particular course of treatment.43 In other cases, when the treatment provided was “not only below the standards of care, but at times grossly improper,” courts have denied summary judgment to corrections officials.44
One interesting issue is how deliberate indifference relates to the standard of care that applies outside of prisons. Courts have offered various ways to characterize deliberate indifference, including, for example treatment that falls below a “reasonable standard of care”45 outside of prison or that does not represent “reasonable professional judgment;”46 treatment “so far below the reasonable standard of care,”47 or treatment that is “so far out of bounds that it was blatantly inappropriate or not even based on medical judgment.”48 Still other cases have turned on whether a doctor affirmatively recognized that a particular course of treatment was the standard of care and yet failed to provide it.49 Courts also disagree regarding what role the medical community's opinion should play. While some courts maintain that judges should not “constitutionalize” the standards of professional medical groups,50 others have held that the treatment of incarcerated people “must be evaluated according to professional standards”51 or that care must be “commensurate with modern medical science and of a quality acceptable within prudent professional standards.”52 In short, courts appear to differ considerably in describing what exactly is necessary for a finding of deliberate indifference.
Our review of federal district and circuit court case law brought under 42 U.S.C. § 1983 — the federal statute under which plaintiffs may sue for the deprivation of constitutional rights by government officials — shows that plaintiffs generally do not prevail when the court sees them as disagreeing with the medical treatment provided by prison doctors rather than demonstrating a lack of treatment,53 or when they cannot muster enough evidence — circumstantial or otherwise — to show the requisite knowledge.54 One defense that courts typically reject in these cases, however, is the “cost” defense, whereby corrections officials claim they cannot afford a particular kind of treatment and thus are not required to provide it.55 To the contrary, “inadequate funding can be a basis for a finding of deliberate indifference, to the extent that it is the non-medical reason for a correctional administrator's interference with the care medical providers have deemed necessary.”56
2. What Qualifies as a Serious Medical Need?
The Supreme Court did not rigorously define “serious medical need” in Estelle, except to say that failure to treat it would cause “unnecessary and wanton infliction of pain.”57 Since Estelle, the Court has held that a medical need can be serious without presenting harms that are immediate or inevitable; what is required is a “serious risk of harm.” In Helling v. McKinney, the Court considered whether involuntary exposure to second-hand smoke in a prisoner's cell is a colorable claim under the Eighth Amendment.58 The Ninth Circuit had held that McKinney stated a cause of action by alleging, with scientific evidence, that his secondhand smoke exposure posed an unreasonable risk of harm to his future health.59
The Supreme Court agreed that McKinney stated a claim, despite prison officials' argument that prisoners must be presently suffering to sue.60 The Court explained that it had “great difficulty agreeing that prison authorities may not be deliberately indifferent to an inmate's current health problems but may ignore a condition of confinement that is sure or very likely to cause serious illness and needless suffering the next week or month or year.”61 For claims of future harm, the Supreme Court instructed that lower courts should consider whether the risk is “so grave that it violates contemporary standards of decency to expose anyone to such a risk.”62 Lower federal courts have not ruled out Eighth Amendment suits premised on adverse medical events that happen shortly after release.63
The majority of federal circuits have described a serious medical need as one that “has been diagnosed by a physician as mandating treatment, or one that is so obvious that even a lay person would easily recognize the necessity for a doctor's attention.”64 Other circuits have established factor-based formulations, including: “(1) whether a reasonable doctor or patient would perceive the medical need in question as ‘important and worthy of comment or treatment,’ (2) whether the medical condition significantly affects daily activities, and (3) ‘the existence of chronic and substantial pain.’”65 In general, lower courts consider whether the condition can be treated, the consequences of foregoing treatment, and the likelihood of a favorable outcome from treatment.66
II. The Eighth Amendment Case for MAT in Jails and Prisons
We argue that refusal by correctional administrators to offer MAT as an option for doctors to provide to incarcerated persons suffering from OUD constitutes deliberate indifference to a serious medical need. To be clear, we are not arguing herein that all incarcerated people have a constitutionally-grounded right to receive MAT. Not all have OUD, and among those who do, a doctor may not advise MAT and/or the individual may not want MAT. Here, we argue only that failure on the part of prison administrators to provide access to MAT as a treatment option for OUD is a constitutional violation, since it prevents doctors from using their medical judgment even when they believe MAT is the appropriate treatment and their patient wants MAT.
This argument pertains only to jurisdictions that do not currently allow MAT as a treatment option. In a correctional facility that has already set up MAT as a treatment option, this argument does not apply. In these facilities, if a prisoner with OUD who wants MAT is denied MAT, he or she may raise an individual claim alleging deliberate indifference to his or her particular circumstances, but this claim would be highly fact specific and is beyond our current scope. To analogize to a different area of prison health care — an argument that it is unconstitutional for a prison to withhold any and all diabetes medications as an option for treatment is different from an argument that a prison violates an individual prisoner's constitutional rights if it provides diabetes medications to some prisoners but fails to provide it to this or that particular prisoner who is presenting with similar symptoms and diagnoses. This paper focuses on the former structure of argument — because the most pressing issue in the current moment is that MAT is not even an option at most prisons.
Our search of case law revealed no cases challenging the failure of correctional administrators to set up a system so that doctors can provide long-term MAT. However, there has been some litigation over the common practice of correctional institutions to discontinue MAT for newly incarcerated people, forcing them to go through withdrawal. Courts have almost universally held the negative effects of opioid withdrawal qualify as a serious medical need that requires treatment (for example, with medications for nausea or with tapered MAT),67 though some have rejected a per se rule.68 Courts have found deliberate indifference to withdrawal symptoms in a variety of circumstances — for example, when officials failed to provide an initial medical examination within a statutorily required amount of time;69 when officials did nothing to treat withdrawal symptoms, despite knowledge of a previous suicide attempt;70 when a doctor ordered no treatment after an individual in withdrawal had a seizure;71 and when an official responded to repeated requests for treatment by saying: “You can stand on your head, tear the place apart, you're not getting methadone.”72
There is, however, a strong theoretical foundation under the Eighth Amendment for giving doctors in correctional institutions the ability to provide MAT even to those individuals with OUD who did not previously receive MAT. In Section A, we briefly examine the medical criteria for diagnosis of OUD and the clinical evidence that long-term MAT decreases illicit opioid use. In Section B, we explain the risk of severe long-term health consequences of untreated OUD — most prominently, deadly overdose and the contraction of dangerous disease — and how these risks affect incarcerated people in particular. We then review a number of studies that have concluded that MAT ameliorates those risks. We argue, in Section C, that on the basis of this evidence, correctional administrators are deliberately indifferent if they refuse, for non-medical reasons, to allow doctors to provide MAT to incarcerated people for whom it is the appropriate treatment. Finally, in Section D we discuss how the opioid crisis has prompted broad support for MAT, for both incarcerated and non-incarcerated individuals. Because the Eighth Amendment question turns, in part, on whether society considers untreated OUD to be a risk “so grave that it violates contemporary standards of decency,” the shift in public and medical support for MAT reinforces the constitutional cruelty of correctional institutions that fail to provide it.73
A. MAT is Effective for Opioid Use Disorder
In what follows, we describe the medical basis of OUD and the effectiveness of MAT in general, moving to prison in the next section. According to the National Institute on Drug Abuse (NIDA), OUD and other substance use disorders are considered “chronic, relapsing brain disease[s].”Reference Volkow74 A diagnosis of severe OUD is made when an individual presents with six or more of the following criteria: 1) use of opioids in hazardous situations, 2) use despite social or interpersonal problems related to use, 3) neglecting major roles to use opioids, 4) physiologic withdrawal, 5) tolerance to opioids, 6) craving for opioids, 7) escalated amount of opioid use, 8) repeated attempts to quit or inability to control use, 9) significant time spent using, 10) physical problems related to use and 11) important activities given up to use.Reference Hasin75
OUD increases the likelihood that users experience sexual violence,Reference Jessell76 depression,Reference Brooner77 maternal and fetal mortality during pregnancy,Reference Winklbaur78 endocarditis, and overdose.Reference Ronan79 It also puts injecting users at high risk of contracting HIV, hepatitis B and hepatitis C.Reference Teesson80 People with OUD are, overall, 20 times more likely to die an early death.Reference Schuckit81
There are several different medications available to treat OUD. Methadone and buprenorphine are considered the standard of care for ongoing treatment.Reference Connock82 MAT is superior to other options for treating OUD, even for those with favorable prognostic factors. For example, a large multicenter, randomized-controlled trial found significantly better results from buprenorphine maintenance than opioid taper. This was true even for patients who only used prescription opioids and who therefore tend to have less injection use, fewer social problems related to opioid use, and a shorter duration of addiction than people dependent on heroin.Reference Moore83 The authors found that only 8.6 percent of individuals were successfully treated — meaning no or minimal illicit use — with two weeks of buprenorphine-naloxone, two weeks of taper, and eight weeks of follow-up counseling.Reference Weiss84 The 91 percent of individuals who tapered unsuccessfully were then offered prolonged access to buprenorphine-naloxone. After 12 weeks of continuous MAT, 50 percent of those same individuals met the criteria for treatment success.85
Research has shown methadone to be similarly effective,Reference Farré86 and numerous studies have replicated these findings.Reference Woody87 While it is difficult to compare the effectiveness of medications for different conditions, the number needed to treat (“NNT”) — meaning the average number of patients who would need to be treated to prevent one bad outcome — is one common standard. In the case of buprenorphine, the NNT to achieve retention in treatment and reduction in illicit opioid use is two.Reference Raleigh88 NNTs for many other psychiatric conditions are higher. For example, the NNT for improved mood for people with depression is six,Reference Arroll89 and the NNT for suppression of heavy drinking among people with alcohol use disorder is 12.Reference Jonas90 When considering non-psychiatric conditions often encountered in prison, such as diabetes and hypertension, the NNT to prevent a macrovascular outcome (such as a heart attack or stroke) is 16 and 54, respectively.Reference Kooy91
The FDA recently approved sustained-release injectable naltrexone, an opioid antagonist that lasts four weeks, as an alternative form of MAT.Reference Lee92 Naltrexone, because it is an antagonist rather than a full or partial agonist, has limited abuse potential. There is preliminary evidence that “extended-release naltrexone [is] as effective as buprenorphine naloxone in maintaining short-term abstinence from heroin and other illicit substances.”Reference Kunøe93 However, other studies suggest that retention in treatment with naltrexone for more vulnerable populations does not approach levels seen with opioid agonists like methadone and buprenorphine.Reference Cousins94
B. The Severe Consequences of Denying MAT to Incarcerated People
Drug use and its corresponding risks is a major issue in correctional institutions. A majority of both state and federal inmates have substance use disorders,95 incarcerated people are ten times more likely to have a history of heroin use,Reference Westerberg96 and rates of injecting drug use in correctional institutions are high (though variable across institutions).Reference Clarke97 Despite the scope of the problem, and the evidence for MAT, people with OUD rarely receive MAT in correctional institutions.98 While the health consequences of OUD on are varied, we focus here on two of the most concrete and severe: death and disease. As MAT reduces the risk of both, failure to establish a system puts lives in danger.
Incarcerated people with OUD are at risk from unintentional overdose both during their term of incarceration and after they leave.99 Prisons, and particularly jails, are not immune to the nationwide overdose epidemic.Reference Noonan100 The most recent data, from 2014, show that alcohol and drug intoxication accounted for 8.5 percent of deaths in local jails101 and 1.4 percent in state prisons.Reference Noonan102 Given the growth of opioid deaths since 2014, and reports of numerous recent overdoses in jails,Reference Johnson103 those numbers will likely continue to increase. By comparison, in 2014, the rate of drug and alcohol poisonings outside of correctional institutions — which includes poisonings as a result of suicide, for example — is 1.88 percent, but it is difficult to compare these rates because of the lack of specificity of death-in-custody reporting.104
The risk of death from unintentional opioid overdose is particularly acute in the days immediately following release from prison or jail. Most incarcerated drug users' tolerance decreases during incarceration — either through effective forced abstinence or because drugs in prison are more difficult to obtain and less pure — which makes overdose more likely after release.Reference Merrall105 This phenomenon has severe consequences. A meta-analysis of six studies from the United States and abroad found that the risk of drug-related deaths after release from incarceration is three to eight times higher in the first two weeks after release than the following ten.106 In Washington State, from 2000-2009, 8.3 percent of all overdose deaths came from the 76,208 people released from incarceration during that time.Reference Binswanger107
Studies of MAT in jails and prisons have shown that MAT can help prevent these deaths. One study in the United States compared post-release behavior between two groups of prisoners who were taking methadone before their incarceration. The group whose treatment was continued had lower rates of drug use and fewer overdoses than the group whose treatment was discontinued.Reference Rich108 In Australia, where MAT is commonly provided in prisons, a study of 16,453 individuals released from prison between 2000 and 2012 found that post-release mortality was almost six times lower for those who had continuously retained in MAT versus those who were forced to withdraw.Reference Gisev109 A study of 12,260 patients from England, where long-term MAT has been provided in all prisons since 2010, showed that participation in MAT during incarceration produced an 85 percent reduction in fatal drug overdoses in the first month after release.Reference Marsden110 Conversely, discontinuing MAT upon incarceration is associated with greater intravenous drug use and needle sharing.111
Another serious and heightened risk for people with OUD who do not receive MAT is the risk of contracting dangerous diseases through needle sharing. The Supreme Court has recognized that prisoners can state a deliberate indifference claim if they are at high risk of being infected with “infectious maladies such as hepatitis and venereal disease …. [e]ven though it was not alleged that the likely harm would occur immediately and even though the possible infection might not affect all of those exposed.”112 Most recently, in discussing the California prison health care system's constitutional deficiencies, the Court focused heavily on how overcrowding “increased the incidence of infectious disease” and how inadequate care put “all prisoners in California” at risk, even those with no present illness.113
Communicable diseases are of particular concern for people with OUD, many of whom are intravenous drug users.Reference Beckman114 As on example, the rate of hepatitis C among incarcerated people is 17 times higher than the rate for non-incarcerated people.115 New infections of hepatitis C in the United States have tripled as a result of the opioid epidemic, with 3.5 million people are now estimated to have the disease.Reference Scutti116 Hepatitis C is primarily contracted through needles used to inject drugs, the sharing of which is common in correctional settings.117 Research has shown “direct evidence” of ongoing hepatitis C transmission among incarcerated drug users.Reference Bretaña118 The same is largely true of HIV, the prevalence of which is three times greater among incarcerated people than in the non-institutionalized population.Reference Springer119 While the precise date of infection is difficult to ascertain, evidence also suggests transmission of HIV among incarcerated people.Reference Dolan120
MAT in jails and prisons would substantially reduce OUD's attendant risks, and the Supreme Court has recognized “secondary” risks of future harm — for example, the risk of lung cancer that stems from exposure to second hand smoke, and, in this instance, the risk of contracting hepatitis C or HIV that stems from OUD — as a legitimate source of Eighth Amendment scrutiny.121 Meta-analyses have found that MAT decreases risky behavior like intravenous drug use and needle sharing;Reference Fullerton122 that MAT reduces illicit opiate use and HIV risk behaviors;123 and that the reduced likelihood of contracting HIV is “largely attributable to the significant reductions in drug-related risk.”Reference Gowing124 The same finding applies to hepatitis C. A study of heroin users who received MAT while incarcerated found that subjects who stayed in treatment for longer periods were significantly less likely to be infected.Reference Dolan125 The ability to resist intravenous drug use is also medically beneficial, because people whose hepatitis C is cured are less likely to become re-infected through future drug use.126
Some corrections departments have begun pilot programs — often funded by Alkermes, the maker of the brand name version of injectable naltrexone, VivitrolReference Goodnough and Zernike127 — that offer patients a single dose of the drug shortly before release.128 To be clear, we argue that many of the above dangers can only be alleviated through ongoing treatment. While providing MAT shortly prior to release may reduce some of the risk of overdose, it would not decrease the spread of disease or the other consequences of drug use within correctional institutions. Just as, under the Eighth Amendment, these institutions must provide ongoing treatment for the chronic diseases of hypertension and diabetes, they likewise must establish MAT programs whereby doctors can provide ongoing treatment for OUD.
C. Failure to Offer MAT Meets the Standard of Deliberate Indifference
To provide MAT, prison doctors need institutional support, whether in the form of licensing to establish a methadone program or to prescribe buprenorphine129 or simply with the administrative burden of setting up a dispensing system. By refusing to provide that support and by rendering inaccessible an entire class of treatment, administrators take away the option for doctors in prison to practice medicine in the way they see fit and therefore deny “a class of inmates the type of individualized treatment normally associated with the provision of adequate medical care.”130 To be clear, our argument is not that the Constitution requires unlimited access to MAT to every incarcerated person or even every incarcerated person who requests it. For example, people who are intermittent opioid users who do not fit the criteria for OUD, people with some co-occurring medical conditions,Reference Jones131 and people who use benzodiazepines may be inappropriate candidates for MAT.Reference Reynaud132 Moreover, while any individual may bring an Eighth Amendment suit claiming improper denial of MAT, that is a separate claim from the one we make, which is that if prison doctors are not permitted to provide MAT, they are unable to treat patients according to their medical judgment.
Prison authorities who are sued for failure to failing to establish MAT programs may argue in response that the consequences — for example, unintentional overdose or the contraction of disease — stem from the illegal drug use itself and therefore attenuates causation. However, it is hard to imagine a court ratifying a correctional institution's decision not to treat HIV because it was contracted from sexual activity that ran afoul of prison rules.Reference Lavender133 Similarly, correctional officers have been held liable for deliberate indifference in numerous cases because they ignored threats or signs of suicide, even though suicide is against prison rules.134
Prison officials may also invoke administrative defenses — most prominently, a preference against MAT, cost, and security. While courts typically defer to correctional officials, they cannot ignore constitutional violations “simply because a remedy would involve intrusion into the realm of prison administration.”135 We discuss the inadequacies of each argument in turn.
Prison officials who are wary of MAT programs do not base their opposition to MAT on medical science. A survey of prison administrators nationwide found the most common reason for not offering MAT has little to do with its effectiveness. Rather, when asked why they did not set up methadone or buprenorphine programs so that doctors could provide MAT, the most common response by officials was that their facility favors “drug-free” treatment.136
Cost is likewise not a defense to constitutional violations, even if the inability to pay is the result of legislative underfunding rather than an internal prison decision. As an example, a federal district court in Florida recently entered a preliminary injunction against the state prison system, giving it two weeks to develop a policy to treat thousands of prisoners with hepatitis C in spite of the cost of $37,000 per patient and the risk that patients could contract the disease again if drug use continues. The judge rejected prison administrators' attempt to claim cost as a defense.
A medical decision premised on an administrative preference is not a decision based on professional medical judgment, and therefore does not receive the same deference as a decision made by a medical provider.137 The Seventh Circuit, for example, defines deliberate indifference as a decision that is “such a substantial departure from accepted professional judgment, practice, or standards, as to demonstrate that the person responsible actually did not base the decision on such a judgment.”138 In other words, “implicit in this deference to prison medical authorities is the assumption that such an informed judgment has, in fact, been made.”139 Deliberate indifference is more obvious if one has access only to doctors who cannot (rather than fail to) provide the standard of care.140 As the Ninth Circuit has held, “access to the medical staff has no meaning if the medical staff is not competent to deal with the prisoners' problems … It must be able to treat medical problems or to refer prisoners to others who can.”141
Cost is likewise not a defense to constitutional violations, even if the inability to pay is the result of legislative underfunding rather than an internal prison decision.142 As an example, a federal district court in Florida recently entered a preliminary injunction against the state prison system, giving it two weeks to develop a policy to treat thousands of prisoners with hepatitis C in spite of the cost of $37,000 per patient and the risk that patients could contract the disease again if drug use continues.143 The judge rejected prison administrators' attempt to claim cost as a defense.144 In any event, cost is not a significant obstacle in the case of methadone. Daily methadone treatment itself (apart from the costs of provision) costs about $30 per month per patient,Reference Donaher and Welsh145 and even with the added costs of provision and counseling at clinics, treatment is around $4,000 annually — far less than treatment for hepatitis C.Reference Rich146
Security concerns — primarily diversion of medication to incarcerated people for whom it is not prescribed — are similarly difficult defenses, particularly because they have proven to be less severe than many corrections officials thought. In a study of buprenorphine and methadone provision at Rikers Island in New York City, relatively few participants were suspected of attempted diversion.Reference Magura147 Methadone, particularly when ingested as a liquid that a nurse can watch patients swallow, is difficult to divert,148 and attempts to do so are infrequent.Reference Gordon149
Prisons and jails should of course have discretion to set up programs in ways that reduce diversion, but refusing to provide this medication simply out of fear of diversion is not constitutionally defensible. Similarly, while there are administrative burdens in providing strong painkillers for cancer patients in prison, patients do receive those medications despite the risk of diversion (though at lower rates than in the outside world).Reference Lin and Matthew150 Anecdotally, officials in Rhode Island and Connecticut, both of which have pioneered MAT programs in their systems, have said that the programs have not increased security concerns.Reference Freyer151
Finally, it is not a defense that a correctional institution offers other non-medical “treatments” to those with OUD. While some prisons offer no treatment for OUD whatsoever, others offer psychosocial therapies — mostly in groups, due to the cost of individualized therapy — which is not as effective as MAT.Reference Farabee and el-Guebaly152 Some studies have found that counseling by itself does not have any effect on illicit opioid use,153 while others have found that the rate and intensity of counseling while on MAT has no effect on treatment retention or use of illicit opioids.Reference Schwartz154 Thus, while psychosocial therapies may be a compliment to MAT, the evidence strongly indicates that MAT is in fact the foundational treatment for OUD.Reference Friedmann and Schwartz155 Psychosocial treatment alone is not sufficient.
D. Evolving Standards of Decency Surrounding MAT and OUD
In Helling, the Supreme Court said that incarcerated people must not only provide medical evidence for the seriousness of their claims, but must also show that “society considers the risk that the prisoner complains of to be so grave that it violates contemporary standards of decency.”156 Since the test for cruel and unusual punishment is dynamic — it “draw[s] its meaning from the evolving standards of decency that mark the progress of a maturing society.”157 Evolving standards of decency have increased scrutiny on other conditions that a decade ago would likely not have been considered serious. Recently, courts “have shown a willingness to look outside ‘traditional’ models of illness in defining medical needs,” for example protecting the rights of a transgender woman forced to terminate hormone treatment, inmates who engage in self-harm and inmates who show suicidal risk factors.158 As one court recently put it, “the proposition that the constitutional minimum with respect to health care has increased over time should be an entirely uncontroversial one; courts find Eighth Amendment violations based on the denial of sorts of care that did not exist decades earlier.”159
Strong evidence has recently emerged of a broad, interdisciplinary, and bipartisan change in views of how people with OUD should be treated. While not universally accepted, the understanding of addiction as a disease has helped to destigmatize addiction. The birth of addiction medicine as a specialty in the medical field has trained a growing number of physicians to treat people with OUD and other substance use disorders.Reference Smith160 These evolutions have changed societal norms, and we argue that there is now a shared understanding that is cruel to deprive a person with OUD the opportunity to be treated by a doctor who can offer MAT as an option.
The current consensus of the medical community — both clinical and academic — is that MAT is the standard of care for treating OUD. A broad range of mainstream organizations — the World Health Organization (WHO),161 the National Institute on Drug Abuse,162 the Substance Abuse and Mental Health Services Administration,163 the Center for Disease Control,164 the American Medical Association (AMA),165 and even the American Academy of Pediatrics166 — all recommend MAT. The WHO goes further and explicitly recommends MAT during incarceration,167 as does the American Psychiatric Association.168
The Affordable Care Act has also raised health care standards. Under current Eighth Amendment jurisprudence, health care available to non-prisoners has become the “yardstick against which the health care status of prisoners is measured.”Reference Malave169 That yardstick is generally taken to be the standard of health care available to poor people, and the Affordable Care Act has moved the stick by providing subsidies to obtain health insurance and encouraging states to expand Medicaid, the baseline level of health care provided to the poorest Americans.170 All state Medicaid programs cover some form of MAT — buprenorphine, methadone, naltrexone, or extend-release injectable naltrexone171 — though in reality, there can be burdensome “fail first” requirements, required preauthorization, limits on lifetime utilization, and dosage limits.Reference Volkow172 In states that have expanded Medicaid, however, there have been substantial increases in the number of Medicaid-covered buprenorphine prescriptions and in prescriptions per enrollee.Reference Clemans-Cope173 Though the data cannot distinguish prescriptions written for newly-covered patients from increased prescriptions for existing patients, it does “suggests a large-scale expansion of access to effective treatment in those states.”174
The response to the opioid epidemic has been strong and bipartisan. President Trump declared the crisis a national health emergency, and Reference Merica175 in 2017 the President's Commission on Combating Drug Addiction and the Opioid Crisis called for offering MAT in jails and prisons in its final report.176 The previous year, Surgeon General of the United States released a 400-page report on addiction, which discussed MAT extensively.177 The governors of 46 states have signed a national Compact to Fight Opioid Addiction, which endorsed MAT and aimed to reduce barriers to access for people on Medicaid.178
Organizations that were not previously supportive, including the National Association of Drug Court Professionals, have come out in support of MAT.179 The same is true of the Hazelden Betty Ford Foundation, one of the largest drug treatment providers in the country, which five years ago reversed course and opted to provide MAT instead of focusing on twelve-step programs.Reference Lopez180
The opioid epidemic claims the lives of 142 Americans on an average day.Reference Donnelly181 The extent of this crisis is changing minds and has reduced the stigma around OUD, pushing MAT into the cultural and political mainstream, and even, as we discuss next, into a growing number of correctional facilities. Contemporary standards of decency demand more of prisons and jails, and some are stepping up to the task.
III. Long-Term MAT in Jails and Prisons Already Exists in Some Jurisdictions
A symbiotic relationship exists between progress through the political system and progress through the courts. As more states implement MAT programs, they serve as a useful proof of concept. Several jurisdictions have now shown that MAT programs are feasible in correctional settings. While the MAT program on Rikers Island began in 1987, only recently are other jurisdictions beginning to follow suit.Reference Vestal182 Connecticut and Vermont have now had pilot programs for several years.Reference Chen and Menard183 Massachusetts appears poised to pass legislation phasing in a statewide program over the next four years.Reference Freyer184 The Federal Bureau of Prisons has also launched MAT field trials in three prisons, treating a small number of patients.185 We focus here on Rhode Island, which is the only state in the country with a comprehensive MAT program for its entire correctional system.Reference Joseph186 Spearheaded by the state's governor, the program was established in 2016 as part of a package of reforms to address the opioid epidemic.Reference Raimondo187
A. A Case Study of Rhode Island
Rhode Island Governor Gina M. Raimondo took office in January 2015, following a campaign which had not focused on the opioid epidemic.Reference Anderson, Pina and Bogdan188 Soon after taking office, however, it became clear that addressing the opioid epidemic would need to be one of her top priorities.Reference Raimondo189 In 2014, 239 Rhode Islanders died from overdose, more than the number of homicides, motor vehicle accidents, and suicides combined.190 Worse, Rhode Island had some of the worst underlying trends in the region. In 2013, the state had the highest rate of illicit drug use in the country, and the highest rate of drug overdose deaths in New England.191
In August 2015, Gov. Raimondo issued an executive order creating the state's Overdose Prevention and Intervention Task Force.Reference Raimondo192 The Task Force was co-led by the directors of the state's department of health and department of behavioral healthcare, and included representatives from the Medicaid office, state and local law enforcement, corrections, the recovery community, and legislative leadership.193 Academic experts in addiction supported the task force's work.194 The executive order charged the task force with developing a goal for reducing overdose deaths and recommendations to achieve that goal by November.195
The Task Force's final recommendations included strong support for a comprehensive corrections-based MAT program.196 The report also recommended the creation of Medicaid-funded opioid treatment centers of excellence, which could serve both as induction points for patients and training grounds for providers,197 and advocated eliminating “fail first” policies from all health insurance policies in the state, both Medicaid and private.198 Together, these recommendations were designed to create comprehensive access to MAT in the state.199
Gov. Raimondo supported the recommendations and submitted a legislative package to the General Assembly for the 2016 session.200 Despite a structural budget deficit of $50 million, Gov. Raimondo found offsetting savings elsewhere in the state budget to support $4 million in new state revenue to support the initiatives, about $2.5 million of which was earmarked to fund prison-based MAT.201 With the offsetting savings already identified, the reforms easily passed the state legislature and the MAT program started the following year.202
Following passage, the Rhode Island Department of Corrections had to design and implement the MAT program. In general, the major design decisions for a corrections-based MAT program are who will be eligible, which medications will be available and in which facilities, and whether induction will be available or only maintenance for those already on MAT upon incarceration.203 For each decision, options which make the program more effective also tend to increase cost or security concerns.204
A comprehensive program offers patients a choice of all three medications: methadone, buprenorphine, and injectable naltrexone.205 Likewise, a comprehensive program would make MAT available to all medically qualifying prisoners, regardless of the facility in which they are incarcerated.206 Several states, such as Massachusetts and Connecticut, have MAT pilot programs in a limited set of facilities, which reduces the risk of diversion, but requires some patients to go without treatment.207 Further, unlike in Rhode Island, many states have split correctional systems, with a combination of state-run prisons and locally-run jails.Reference Kaeble, Glaze, Tsoutis and Minton208 Imposing MAT programs on local jails may present additional funding and political challenges not present in state prisons.
Deciding whether to offer induction — meaning whether or not to start people on MAT in prison — is a significant choice. Induction is the most difficult part of MAT, and generally requires physician oversight.209 Arranging for a physician to be present in the prison is both difficult for corrections officials and expensive.210 On the other hand, programs that do not offer induction cannot serve patients who are using opioids when they are incarcerated but have not yet started treatment, which is a significant portion of the population that would benefit from treatment.211
Finally, programs must choose how to handle individuals with very long or short sentences. A comprehensive program would make treatment available to anyone who is medically qualified.212 To reduce expenses, however, correctional facilities may want to treat people with long sentences differently from people with shorter sentences.213 People with longer sentences may be tapered off treatment when they enter a facility and brought back on treatment shortly before leaving, whereas people with shorter sentences would be continued on treatment during the entirety of their incarceration. People with very short term sentences, moreover, may not be offered induction because they might leave before induction is complete.214
Outcomes are being measured and evaluated as this article goes to press. On February 14, 2018, researchers published an initial evaluation of the program, comparing results from the six months of the program to the period one year prior. As expected, the number of inmates receiving MAT in a given month grew significantly, from 80 inmates pre-implementation to over 300 inmates afterward. Of inmates on treatment, 59 percent received methadone, 39 percent received buprenorphine, and the remainder received naltrexone. The number of individuals dying of overdose within six months of release fell by 61 percent, from 26 deaths to 9 deaths, a statistically significant result. While these results are preliminary and did not take place in the context of a randomized-controlled trial, the reduction in deaths is remarkable and points to the large potential impact of MAT programs.
In Rhode Island, as noted, the MAT program was designed to be comprehensive.215 The program makes all three forms of medication available, and is offered in all jail and prison facilities statewide. Induction is available, and most prisoners are clinically eligible, with treatment for long-term prisoners tapered off and restarted, if medically appropriate, shortly before release. Rhode Island runs a unified prison system, with all facilities located on one central campus and under the state's direct control, which made the program easier to introduce.216 In total, the program was projected to provide treatment for around 2,200 prisoners.217 Two-thirds of those patients are expected to start treatment in prison, with the last third already having started treatment in the community.218
Outcomes are being measured and evaluated as this article goes to press. On February 14, 2018, researchers published an initial evaluation of the program, comparing results from the six months of the program to the period one year prior.219 As expected, the number of inmates receiving MAT in a given month grew significantly, from 80 inmates pre-implementation to over 300 inmates afterward. Of inmates on treatment, 59 percent received methadone, 39 percent received buprenorphine, and the remainder received naltrexone. The number of individuals dying of overdose within six months of release fell by 61 percent, from 26 deaths to 9 deaths, a statistically significant result.220 While these results are preliminary and did not take place in the context of a randomized-controlled trial, the reduction in deaths is remarkable and points to the large potential impact of MAT programs.
IV. Conclusion
The opioid crisis claims the lives of tens of thousands of people every year, many of whom cycled through cells in jails and prisons across the country before an overdose or chronic disease took their life. Our current moment presents the opportunity to change the course of how we treat OUD in our prisons. Today, most prisons and jails force people to suffer withdrawal and put them at risk of life-altering diseases, only to release them back into a world with reduced tolerance, primed for unintentional overdose. In doing so, corrections officials are ignoring the decades-old science on MAT, which has been proven to reduce negative consequences of OUD in prison and in the community. They also ignore evidence from Riker's Island, Connecticut, Rhode Island, and other jurisdictions illustrating that such programs are not beyond the capacity of correctional institutions.
This is not a mere policy decision. Denial of access to MAT for patients who medically require it violates the Eighth Amendment's prohibition of cruel and unusual punishment because it is deliberately indifferent to the serious medical needs that OUD presents. Failure to provide MAT may violate other laws as well — for example, the Americans with Disabilities Act — and these alternatives are worth more scholarly consideration going forward.221 Regardless of whether the courts ultimately agree, there is much to be gained from a political reckoning around how we think about addiction, “addicts” and compassionate medical care for the most vulnerable among us. No one argues that medication-assisted treatment in prisons will end the opioid crisis. But it is a good place to start.
