Ethical and Safe Discharges from Psychiatric Units during the COVID-19 Pandemic

Mental Health Law

We begin with some of the original court cases that pertain to inpatient psychiatric admission and discharge. Wyatt v. Stickney³ (1971), initiated by guardians of patients admitted to Alabama State Mental Hospital, established that inpatient psychiatric units are justified in keeping a patient only while they provide treatment.³ The case clearly states that patients cannot be held for custodial care, as hospitals are not a penitentiary to house people indefinitely.³ Another landmark case, O’Connor v. Donaldson⁴ (1975), adds that hospitals cannot detain patients simply to give them a higher quality of living compared with what is available to them in the community.⁴ The court ruled that “commitment must be justified on the basis of a legitimate state interest” (Ref. 4, p 580) and established that hospitals cannot restrict patients if they can live safely on their own or with family.⁴

An additional case during the pandemic is also relevant. In Foster v. Commissioner of Correction⁵ (2020), the Supreme Judicial Court of Massachusetts ruled that substance use treatment facilities need to provide individuals who are civilly committed for treatment with a new hearing to calculate the risk and benefits of admission during the pandemic.⁵ Part of the ruling argues that civil commitment for substance abuse treatment during the COVID-19 pandemic does not sufficiently advance treatment goals, thereby violating individual’s substantive due process rights.⁵ Although this case relates to substance use treatment and not specifically to an inpatient psychiatric unit, it is an example of how states have contemplated the risks and benefits of inpatient treatment. The ruling suggests that infection control measures in inpatient settings, such as no longer offering groups or milieu treatment, may degrade the ability of a program to offer the same level of care that would otherwise be expected.

While establishing important precedents, these rulings do not definitively address the question of how to proceed if a COVID-19 positive or exposed patient no longer experiences psychiatric symptoms warranting inpatient level of care, yet their shelter, family, or residence will not accept them back due to concern about contagion or inability to allow the space or support needed to isolate or quarantine from others. The addition of COVID-19 illness or exposure to our standard equation of evaluating readiness for discharge adds a new element in defining whether a person can live safely on their own or with family. Patients who would otherwise be considered safe for discharge may have ongoing psychiatric symptoms. If these symptoms interfere with their ability to understand or follow isolation or quarantine recommendations, that may shift the decisional balance toward the patient’s no longer being able to live safely in the community. In these cases, treatment is more obviously active and would aim to target these symptoms until the patient can be determined to be safe. In cases where patients’ symptoms do not significantly interfere with their ability to understand or adhere to isolation or quarantine, it becomes more complicated. The patient still may not have a safe place to go in which to isolate or quarantine, and though continued inpatient care involves ongoing active treatment regardless of patient level of symptomatology, one could argue that the level of treatment available while a patient is under strict isolation for COVID-19 is likely less than that offered to other inpatients who are not under isolation. Hence, there is a distinction between those patients who remain psychiatrically symptomatic and are a danger to themselves or others because of their inability to understand isolation rules, but otherwise are safe for discharge under non-COVID-19 circumstances, and those patients who no longer have psychiatric symptoms and remain a danger simply because they have no safe place to isolate.

In New York City (NYC), one solution to the safe disposition problem came from underutilized hotels that were repurposed as COVID-19 hotels to allow homeless individuals with COVID-19 to complete the quarantine period rather than returning to shelters.⁶ The criteria for this temporary housing were strict and the resource was scarce and often unavailable.⁶ The NYC Department of Homeless Services issued a COVID-19 hospital referral application for isolation sites, however serious mental illness or history of suicide ideation or attempt were considered criteria for exclusion.⁶ These limitations essentially meant that people with serious mental illness could not go to shelters as the shelter system did not want to take on additional liability or responsibility in their already burdened system.

Quarantine Law

The court cases cited above relate to mental hygiene laws rather than quarantine laws.^3,–,5 In a recent article,⁷ Eric Broutman, J.D. explained that many of the theories used in public health law are drawn from mental hygiene law. Both restrict individual liberties based on something that may happen in the future to protect the rights and safety of the community.⁷ Just as with mental hygiene law, in which risk of harm to self or others must be proven for involuntary admission, quarantine law stipulates that the government must prove that the individual poses a danger to others via contagion to invoke involuntary quarantine.⁷ The standard of proof to quarantine an individual would be the same used to involuntarily commit a psychiatric patient, which is clear and convincing evidence.^8,9 Once such danger is proven, federal law permits the government to quarantine individuals in two circumstances: an infected individual can be prevented from entering the country or from crossing state lines.⁷ Although there are federal laws that allow for the quarantine of individuals, they relate mostly to circumstances involving international, or interstate travel.^10,11 The bulk of quarantine law emanates from individual states exercising their police powers and parens patriae powers.¹²

Police powers of states may be used by public health officials to compel treatment, direct or prohibit certain behaviors, or detain or isolate individuals based on the legal principle of salus publica (preventing or avoiding public harm).¹² An early example of state police powers being used to quarantine comes from soon after the end of the Revolutionary War when Philadelphia was isolated to control the spread of yellow fever.¹² During the 1950s and 1960s, state police power (and by extension orders of quarantine) became much more heavily scrutinized and restricted by the Supreme Court in favor of preserving individual rights and freedoms.¹² Each state has public health laws that provide for the control of communicable diseases. In New York State, Article 21 of the Public Health Law lays out the powers and duties of local boards of health and health officers in controlling communicable diseases, which include: medical inspection and regulation of people infected with or exposed to diseases; provision of care and isolation for people with communicable disease in a safe environment to protect the public; and prevention of all interactions with infected places and people, including provision for purification and cleansing if necessary.¹³

On March 3, 2020, the Westchester County Health Commissioner used this power when she ordered the closure of Young Israel of New Rochelle synagogue and the self-quarantine until at least March 8 for congregants of the synagogue who attended services or events on February 22 and 23, 2020. The Health Commissioner warned that those who did not self-quarantine would be mandated to do so by the County Department of Health.¹⁴ In New York, the state Public Health Law goes further in Section 2120 to provide the option for a physician to notify a health officer when a person afflicted with a communicable disease is unable or unwilling to prevent danger of infection to family or other contacts.¹⁵ The health officer then investigates the situation and can refer the situation to a magistrate who may then commit the individual to any hospital or institution established for the care of people suffering from such a communicable disease on the basis of danger to others.¹⁵

Interestingly, New York City has its own set of guidelines given the city’s density. An appointed commissioner in New York City is authorized to detain suspected infectious individuals for up to 60 days without a court order in a hospital or their home; however, individuals have a right to a hearing.⁷ A notable example of a federal quarantine order occurred in 2007 when the Centers for Disease Control (CDC) required an individual, Andrew Speaker, to be quarantined in a New York City hospital based on suspected drug-resistant tuberculosis.¹⁶

The distinction between isolation and quarantine is important. Quarantine pertains to containing individuals who have had exposure and are deemed to be potentially contagious, while isolation applies to individuals with known infections.¹⁶ Although distinct, both quarantine and isolation procedures are often influenced by religious, political, and economic biases.¹⁶

Precedents from Other Epidemics

The literature regarding notable infectious disease epidemics, such as with human immunodeficiency virus (HIV) and tuberculosis (TB), provides a framework for how to grapple with the ethics strains of discharging psychiatric patients during uncertain times. In the 1980s, physicians who discharged HIV positive patients were conscious of the potential harm patients could inflict on others, the discrimination they would experience, as well as the lack of housing options for them.¹⁷ At that time, experts concluded that state hospitals should not serve as penitentiaries for competent and stable HIV positive patients, even if they have the potential to spread the virus.¹⁷ One could infer from this logic that inpatient psychiatric units should not be used as isolation centers, even during infectious disease outbreaks, and, by extension, difficulty with disposition placement may not be enough to justify an extended stay on a psychiatric unit.

In contrast to guidance provided during the HIV epidemic, however, select literature related to tuberculosis offers a more cautious route with more lenience toward holding patients while implementing safeguards to promote containment in vulnerable populations. Specifically, an emphasis is placed on the necessity of comprehensive discharge planning to maintain tuberculosis treatment compliance during management of outbreaks. Bayer et al. posit that effective discharge planning includes securing a residence with a range of accessible social services and treatment options.¹⁸ They argue that governments are ethically responsible to deliver such resources to facilitate treatment, especially when treatment is obligated.¹⁸ Of note, in the 1990s a select number of nonadherent tuberculosis patients in a direct observed therapy clinic in Denver, Colorado, were sent to locked units to receive treatment to protect public health.¹⁹ Of the 20 patients detained, 18 had alcohol use disorder and two had untreated schizophrenia.¹⁹ Despite comprehensive outpatient treatment, Colorado allowed for even more restrictive measures to contain a tuberculosis outbreak.

Although HIV and TB are both devastating illnesses with high morbidity and mortality, the difference between these two approaches is likely rooted in the difference in transmissibility between the two illnesses. The much higher rate of community TB transmission as an airborne pathogen leads to a more restrictive response and is a better analogy for the current SARS-CoV-2 pandemic. The examples from HIV and TB lead us to conclude that it is the responsibility of the provider, as well as the state, to ensure that patients receive comprehensive treatment resources and safe discharge plans, including provisions to protect public health, and that the level of infringement on patient rights warranted is likely proportional to the level of transmissibility, morbidity, and mortality of the infectious illness.

Ethics and Professional Considerations

Ethics concerns about safe discharge planning existed before COVID-19, however the pandemic has highlighted the intensity of these decisions. Hospitals need to calculate the overall benefit and risk of admission for patients during a pandemic. Inpatient psychiatric hospitalization could create a potential exposure for patients as they are in closer contact with others than they may otherwise be in the community, and often in contact with others who may have severe psychiatric symptoms which preclude their ability to comply with social distancing, mask-wearing, or even basic hygiene. In addition, given that much of the therapeutic milieu, such as groups and visiting, have been paused or modified significantly, patients may not receive the accepted level of care that was provided before the pandemic. Sorrentino et al. argue that without the full support of the multidisciplinary team, discharges may be further delayed during COVID-19, as there are fewer personnel evaluating patient progress and fewer therapeutic interventions available.²⁰ The advantage of admission is further weighed against patient autonomy, beneficence, nonmaleficence, and responsibility to the community.²⁰

Ghossoub and Newman explore the responsibility psychiatrists must consider when determining the safety of third parties who may contract the infection from patients.²¹ Patients with serious mental illness may have poor insight or judgment about their COVID-19 status that may magnify their risk of transmission.²⁰ Psychiatrists need to use clinical judgment and legal consultations to calculate their duties to their patients and communities.²¹ The American Psychiatric Association (APA) ethics committee has specified that caregiver safety is an important consideration in discharge planning.²² They cite principles including fidelity, nonmaleficence, and proper distribution of scarce resources in their reasoning and posit that discharging infectious patients to the street poses a safety risk both to the patient and the community. The APA ethics committee stated that, in the event of a public health emergency, involuntary patients can be quarantined if the process carefully discerns between seclusion for behavioral reasons and infectious ones. They defend this statement based on protection or preservation of self and the community. In June 2020, the APA Board of Trustees published guidance on admitting and discharging psychiatric patients during communicable disease outbreaks and explicitly warned against premature discharge as it can result in adverse outcomes for patients and the community.²³ They emphasized the need to address patients’ mental health needs while also considering the potential risks of the communicable disease at different levels of care and stated that delivery of safe and appropriate treatment at all levels includes adequate protection from infection.²³ Although the statement does not specify COVID-19 infection status, the overall tone favors inpatient admission as long as safety measures and potential risks are considered.

In addition to the potential risks outlined above, there is a risk of financial harm to patients by holding them longer in the hospital. With respect to hospital billing, it would be untenable for patients to bear the cost; rather, we anticipate that coverage would be discussed with third payers as usual. As an example, in March 2020, MassHealth issued a statement that acute inpatient hospitals could bill MassHealth for patients no longer requiring an inpatient level of care but who require quarantine in the hospital or who cannot be safely discharged due to COVID-19 status.²⁴ In instances when insurers are unwilling to cover the costs, medical centers may elect to absorb the costs as they do in other circumstances. Another potential financial harm is related to impeding a patient’s return to gainful employment; however, this impact may be negligible compared with the person’s psychiatric illness and COVID-19 infection or exposure.

Case Scenario

To further explore these ethics questions, we present here a common scenario from an inpatient psychiatry unit in New York City during the initial COVID-19 peak in the United States. John Doe is a middle-aged man with a history of schizophrenia who resides with his family, some of whom are elderly. Mr. Doe presented to the hospital with worsening symptoms, including aggression and violence in the context of medication nonadherence. During a prolonged hospitalization, he was eventually transitioned to clozapine for treatment-resistant psychosis with gradual improvement. With reduction in symptoms and greater ability to care for himself, he was determined to be psychiatrically stable and no longer to pose a danger to himself or others.

As his treatment team worked on disposition plans, including helping his family feel safe with him returning home, he tested positive for COVID-19, although he remained asymptomatic. His family were concerned about having him return home to self-isolate given the risk of transmission.

At this point in Spring 2020, COVID-19 rates were surging toward their peak in New York City, fear was overwhelming and, despite guidance from the CDC about self-isolation, housing options and shelters were cautious and did not yet have clear policies on accepting people who had tested positive. In view of Mr. Doe’s new COVID-19 positive status, the New York City centralized shelter system predicted it would take weeks to find him a bed.

Mr. Doe was being held on an inpatient psychiatric unit involuntarily and had been determined to no longer meet criteria for that legal status based on improvement in his psychiatric condition prior to testing positive for COVID-19. He requested to be discharged. The team now had to reconsider his capacity within this new circumstance, and Mr. Doe had a tenuous grasp of his COVID-19 diagnosis and its multiple implications.

As we consider this case in an ethics framework, we consider our obligations to Mr. Doe. We have a duty to honor his autonomy by treating him in the least restrictive setting. There is a fine balance between protecting public health and maintaining individual privacy and liberty. Childress et al. offer five justifications for overriding individual liberty to promote public health policies, which include the anticipated effectiveness of policies that breach moral the guidelines, the proportionality or balance of potential benefits versus risks, the necessity of the policy to achieve public health measures, minimized infringement on moral considerations, and public justification to relevant parties.¹

If we were to honor Mr. Doe’s desire to be discharged without a safe disposition, he would likely either return to his family’s home, possibly endangering them by exposing them to COVID-19, or end up on the street where he would be unlikely or unable to self-isolate and thereby potentially spread the virus to others. Based on the precedent from O’Connor v. Donaldson⁴, he could not be held in the hospital for custodial care if he was in fact no longer receiving acute inpatient level of treatment, and the hospital and state would thus have a responsibility to provide for housing or respite.⁴ As an alternative to this, quarantine law in NYC would allow the commissioner to detain him in the hospital to serve out his quarantine period.^7,16

Mr. Doe’s case highlights the rigid requirements for discharge despite alternatives which are entirely unsatisfactory and potentially dangerous. Arguably, Mr. Doe was still a danger to others given his unknown contagion status. Furthermore, he might even be considered a danger to himself because if he were to infect his own family he could theoretically disable his support system and be left unable to care for himself. Accordingly, an important question to ask in these situations is whether the danger the patient presents is an outgrowth of his mental illness or purely based upon his positive COVID-19 status. Under the former, the patient may remain hospitalized under current mental health laws, the latter would require the utilization of quarantine laws.

Ultimately, Mr. Doe agreed to work with the treatment team to find a safe disposition plan but declined to convert to voluntary legal status. He continued to be treated on the inpatient psychiatric unit, which out of necessity at the time became a designated COVID-19 positive psychiatric unit within a medical hospital system that was overwhelmed with patients with COVID-19.² His team continued to target his impulsivity, self-care ability, insight, and judgment, which were all felt to be affected directly by his psychiatric symptoms related to his diagnosis of schizophrenia. Shelter housing continued to be unavailable. Regular contact was maintained with Mr. Doe’s family via phone as he completed the recommended COVID-19 isolation period. With much encouragement, support, education and further family therapy interventions, family members were able to feel safer accepting Mr. Doe back, leading to his much anticipated and long-overdue discharge.