For healthcare professionals only


Examining the role of healthcare professionals in the use of solitary confinement

BMJ 2017; 359 doi: (Published 24 October 2017) Cite this as: BMJ 2017;359:j4657
  1. Cyrus Ahalt, co-director,
  2. Alex Rothman, associate director of policy,
  3. Brie A Williams, founding director
  1. Criminal Justice and Health Program, University of California, San Francisco, Department of Medicine, Division of Geriatrics, 3333 California Street Suite 380, San Francisco, CA 94118, USA
  1. Correspondence to: B Williams brie.williams{at}

Cyrus Ahalt and colleagues explore the conflicting responsibilities of healthcare professionals involved in solitary confinement, a correctional practice that persists in prisons around the world despite a growing body of evidence describing its harmful effects

Key messages

  • The United Nations defines solitary confinement exceeding 15 days as torture, reflecting mounting evidence of associated psychological and physiological harms

  • Healthcare professionals who work in prisons face unique challenges for which they often lack adequate guidance and support

  • Professional medical organisations, such as the World Medical Association and the World Health Organization, and international bodies, such as the United Nations, should lead reform of prison health systems and support healthcare providers practising in these settings

  • Promoting the translation of standards from community care to prison health systems may reduce dual loyalty concerns among healthcare providers

  • Prison healthcare providers should be supported by training in medical ethics

Prison healthcare professionals work in a unique clinical environment designed to punish rather than to heal.123 Amid global calls for penal reform, healthcare professionals have an ethical responsibility4 to speak out about correctional practices that endanger health and human rights.5 We examine this responsibility in relation to prolonged solitary confinement, a practice that persists in prisons around the world.

Prolonged solitary confinement is widespread

Juan Méndez, the United Nations special rapporteur on torture, has defined solitary confinement lasting longer than 15 days as torture,6 a finding subsequently incorporated into the UN’s minimum rules for the treatment of prisoners, also referred to as the Mandela rules. These rules require that solitary confinement be used only in exceptional circumstances, as a last resort, and for the shortest amount of time possible.7 Yet solitary confinement of longer than 15 days remains in widespread use around the world,89 including in Scandinavian nations often described as paragons of progressive penal reform1011 and even for prisoners who are on remand—incarcerated while awaiting or during their trial.12

Reform of solitary confinement is complicated by it typically being applied at the discretion of prison administrators without due process or access to appeal.13 Mounting evidence of serious mental health effects associated with prolonged solitary confinement,14 however, has led many jurisdictions to limit its use for people with pre-existing mental illness and for those who develop mental illness while in isolation.15

The principles of medical ethics, the Hippocratic oath to “abstain from all intentional wrongdoing and harm,” and the physician’s oath of the Geneva Convention all prohibit healthcare providers from participating in treatment that amounts to torture.41617 The Mandela rules further require that providers pay particular attention to prisoners held in isolation. But many prison healthcare professionals encounter a fundamental tension: the expectation that they assess the health of patients in solitary confinement while knowing that such assessments might be used to extend patients’ exposure to a practice that is known to harm health.

The case of Arthur Johnson

In September 2016, a court in the United States ordered the removal of a prisoner, Arthur Johnson, from solitary confinement, where he had been living for 36 years.18 Despite only one minor behavioural infraction since 1987, Johnson was kept in indefinite isolation, confined to a small cell with lights that remained on for prolonged periods. In 2015, Johnson’s correctional mental health provider was asked to conduct a psychological review as part of an annual review of his housing status. Court documents show that the provider “had not developed a communicative relationship with [Johnson] at the time” but was asked to “vote” on whether Johnson should remain in solitary confinement. The provider’s vote to retain him in isolation, she later acknowledged, was “based entirely on decades old escape attempts.” Prison officials decided to keep Johnson, then 64 years old, in indefinite isolation based “in large part” on the provider’s assessment.18 Johnson filed a lawsuit contesting his ongoing isolation shortly thereafter.

Craig Haney, a psychologist and expert witness testifying on behalf of Johnson, reviewed Johnson’s treatment files from 2003 through 2013.19 He described the care that Johnson received over those years as “superficial psychological monitoring . . . despite clear, substantiated risks to his psychological wellbeing” and criticised numerous “mental health entries and observations [that] reflect little more than endorsements of corrections-based (rather than psychologically informed) judgments and recommendations.” Haney said that the mental health assessment and care that Johnson received during this decade in solitary confinement consisted predominantly of brief “cell front” checks for suicidal or homicidal ideation conducted by non-health staff (meaning that psychological evaluations were completed through his cell door rather than in a private office). Less frequent assessments by mental health professionals were also primarily brief, cell front interviews. Johnson’s resulting “mental health file” comprised short reports that sometimes repeated the same language verbatim and, on the whole, were more correctional than health oriented. Mental health assessments outside the cell took place “sporadically and infrequently, at several year intervals.” These, by Haney’s estimation, still did not “remotely represent in-depth psychological evaluations, assessments, or analyses” and “concluded with a correctional—not psychological—opinion.”19

Haney held extensive interviews with Johnson in person and found that he was “struggling to maintain his sanity” and “approaching losing the will to live.”19 The court, siding with Haney’s assessment, found that Johnson’s decades in isolation had caused “escalating symptoms of mental degradation” and probably inflicted “irreparable harm.” The court ordered his prompt release from solitary confinement to the prison’s general population, where he currently resides.18

Challenges facing prison health professionals

This US case exemplifies the global problems faced by health professionals who treat patients in solitary confinement. Prison health professionals worldwide must provide patients with community standard (“equivalence of”) care in institutions designed to deprive liberty. This arrangement can result in “dual loyalty” challenges: conflicts that arise between providers’ obligations to their patients and their deference, whether explicit or implied, to the correctional institution that bears punitive responsibility for their patients.20

Several international organisations prohibit prison healthcare providers from participating in punitive correctional practices.8212223 Such guidelines, including those of the UN’s Istanbul Protocol22 and the World Medical Association,23 require that correctional healthcare providers attend exclusively to the physical and mental health of prisoners, which includes both the standard duty to care and the responsibility to safeguard prisoners from cruel, inhumane, or degrading treatment. Despite this guidance, ethical challenges related to dual loyalty remain unresolved for many prison healthcare providers, often with dire consequences.2024

The role of healthcare providers in torture at the Guantanamo Bay detention camp—and the related collaboration between the American Psychological Association (APA) and the CIA and the Bush administration—brought renewed attention to dual loyalty concerns in the prison context. The APA’s misplaced loyalty to the state directly undermined the health and human rights of patients at Guantanamo Bay. APA policy at the time allowed psychologists to participate in practices termed “enhanced interrogation” and was used by the US Department of Defence and others to justify, expand, and prolong torture ostensibly approved by experts from healthcare professions.252627 Such complicity represents a profound breach of the physicians’ oath in the Declaration of Geneva, which states: “I will not use my medical knowledge to violate human rights and civil liberties, even under threat.”16

The exceptional case of the APA and Guantanamo Bay received widespread coverage, but the effects of dual loyalty on prisoners in solitary confinement, like Arthur Johnson, are less well documented. Johnson’s mental health providers were directly accountable to prison administrators and, if the court testimony is accurate, adhered to an arbitrary schedule of health assessments determined by Johnson’s correctional status rather than his health. They conducted assessments designed to accomplish a correctional goal (whether the patient could withstand ongoing solitary confinement) rather than a health goal, under conditions (such as brief, cell front interviews) that fall far short of equivalent standards of community care.528 These providers witnessed conditions (such as lights that never turned off) and recorded outcomes of even brief assessments (such as “non-communicative”) that should have triggered efforts to remove Johnson from isolation but did not. Ultimately, his healthcare providers gave a clinical endorsement of Johnson’s mental fitness for indefinite isolation, violating ethical proscriptions against participation in punishment, particularly forms amounting to torture.

Clinicians in community hospital settings, on the other hand, are routinely asked to assess the appropriateness of involuntary psychiatric holds. In such cases, clinicians must undertake a robust assessment, including of a patient’s immediate threat to their self and others. Only a clear, documented risk of immediate physical harm can result in the deprivation of liberty. The involuntary hold is then imposed only as long as the immediate risk to health and safety persists.29 This community standard mirrors standards for the ethical use of solitary confinement advocated by the UN special rapporteur on torture,6 the Mandela rules,5 the American Bar Association,30 and the Istanbul Protocols22: that it be used only in exceptional circumstances, as a last resort, and for as short a time as possible. But as Johnson’s case shows, and others have described,203132 prison health providers need support beyond appropriate guidelines to achieve an ethical standard of care for their isolated patients.

Opportunity for leadership

Many of the reforms needed to make criminal justice systems safer and more effective, from improving police interactions with mentally ill people to optimising transitional healthcare to the community, require leadership from the healthcare professions. Enhanced leadership and action is particularly critical in eliminating prolonged solitary confinement. Unlike other pressing matters in penal reform at the intersection of human rights and medical ethics—including capital punishment and forced feeding3334—very few experts are calling for the absolute prohibition of solitary confinement. Rather, its use for short periods is generally viewed as a necessary, temporary correctional tool to de-escalate dangerous or potentially dangerous situations. This places even greater importance on the role of health providers to ensure that solitary confinement is used for the shortest amount of time possible, consistent with prisoners’ health and human rights.

Statements by a growing number of international bodies have created a foundation for physician led reform of solitary confinement. The US National Commission on Correctional Healthcare states that healthcare professionals “should not condone or participate in cruel, inhumane, or degrading treatment”35 and the Council of Europe requires that medical practitioners inform the prison director when a prisoner’s physical or mental health is put at risk by solitary confinement (rule 43.3).36 But a broader coalition of medical leaders must provide greater leadership to ensure that correctional healthcare providers have the specific guidance, training, supervision, and support they need to assess, treat, and advocate for their patients in solitary confinement (box 1).45

Box 1: Supporting the needs of prison healthcare providers

  • Global bodies could link the Mandela rules to the principle of equivalence in care more explicitly by calling for community hospital standards governing the use of mandatory psychiatric holds to be extended to solitary confinement

  • Medical associations could issue guidance on minimum professional standards for health assessment in correctional settings—such as confidential, therapeutic environments and a required disclaimer on all evaluations of isolated patients that such conditions pose health hazards

  • Professional societies and academics could partner to develop and disseminate curriculums for correctional health providers covering human rights, the principles of medical ethics related to the protection of prisoners, the health risks associated with prolonged solitary confinement, and dual loyalty. This information is currently available in the World Health Organization’s Prisons and Health.3738

  • As advocated by the UN Office on Drugs and Crime, the World Health Organization, and others,394041 medical associations could support jurisdictions where correctional healthcare systems are under correctional authority (such as the Ministry of Justice) to transition to the responsibility of government run health agencies (such as the Ministry of Health) This has been successful in the UK, Norway, and elsewhere.2042

  • Expanding on the model of the European Committee for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment,43 medical associations and government health agencies could implement a system of external oversight to prevent healthcare providers from participating in prolonged solitary confinement, to impose appropriate penalties, such as the loss of licensure, when such participation occurs,44 and to advance whistleblower protections for prison staff who report violations of validated code of medical ethics violations.

In Arthur Johnson’s case, and in countless other examples of prolonged solitary confinement throughout the world, the medical profession has abdicated to their legal and correctional colleagues the responsibility to ensure that incarcerated individuals are protected from cruel, inhumane and degrading treatment and receive quality, community-standard healthcare. It is time for the medical profession to claim a larger leadership role in criminal justice and penal reform. We can start with an effort, beyond guidelines and statements, to change the role that clinicians play in medically harmful correctional practices such as solitary confinement.


  • Contributors and sources: All authors meet all four criteria for authorship in the ICMJE Recommendations. CA, AR, and BW developed the concept for this manuscript. CA drafted the manuscript. AR and BW reviewed and revised the manuscript. No others contributed to this manuscript in any way. CA is the guarantor of the article.

  • Competing interests: We have read and understood BMJ policy on declaration of interests and declare the following interests: BW has served as an expert witness and as a court consultant in legal cases related to prison conditions of confinement. These relationships have included: National American Civil Liberties Union; Squire Patton Boggs; the Center for Constitutional Rights; the Disability Rights Legal Center; the University of Denver Student Law Office; Farris, Vaughan, Wills and Murphy LLP; and Rosen Bien Galven and Grunfeld LLP. No organisation listed here played a role in the design, recruitment, data collection, analysis or preparation of this manuscript. BW and CA received salary support from the Jacob and Valeria Langeloth Foundation while contributing to this manuscript. The funder played no role in the analysis or preparation of this manuscript.

  • Provenance and peer review: Not commissioned; externally peer reviewed.


View Abstract