The current coronavirus disease 2019 (COVID-19) pandemic has triggered an intense global research effort to inform the life-saving work of frontline clinicians who need reliable information as soon as possible. Yet research done in pressured circumstances can lead to ethical dilemmas, especially for vulnerable research subjects. We present the case of a child with neurocognitive impairment who is diagnosed with COVID-19 infection after presenting with fever and a seizure. The child lives in a group home and is in the custody of the state; her parents lost parental rights many years ago. Some members of the health care team want to enroll her in a randomized clinical trial evaluating an experimental treatment of COVID-19. For minor patients to enroll in this clinical trial, the institutional review board requires assent of patients and consent of guardians. An ethics consult is called to help identify relevant concerns in enrollment. In the accompanying case discussion, we address historical perspectives on research involving people with disabilities; proper management of research participation for people with disabilities including consent by proxy, therapeutic misconception, and other threats to the ethical validity of clinical trials; and the potentially conflicting obligations of researchers and clinicians.

A 12-year-old girl is admitted to the hospital with status epilepticus and a fever. She has a history of epilepsy for which she takes topiramate and clobazam. She also has severe neurocognitive impairment secondary to hypoxic ischemic encephalopathy; she requires assistance with all activities of daily living, and her only consistent form of communication is that she cries when in pain. She tests positive for coronavirus disease 2019 (COVID-19). Her seizures abate after receiving antiepileptic medications and, after 12 hours, she is clinically stable and appears to be at her mental status baseline.

On the basis of her presentation, the girl meets criteria for a randomized trial comparing a new antiviral drug targeting COVID-19 to placebo. To be enrolled, patients must be 12 years old or older, positive for severe acute respiratory syndrome coronavirus 2, hospitalized, and have oxygen saturations >94% with radiographic evidence of pulmonary infiltrates. This patient did not have pulmonary symptoms or hypoxemia but did have an abnormal chest radiograph, thereby qualifying her for inclusion. Given the severity of this patient’s initial presentation, her underlying medical condition, and uncertainty around COVID-19 prognosis in this population, the girl’s care team considers enrollment.

The girl resides at a long-term care facility, having been removed from the care of her biological parents at age 7 because of concerns for neglect. Parental rights were terminated at that time. She is in the custody of the Department of Children and Families, which also maintains medical decision-making rights and the right to give permission for her participation in research. They are willing to give permission for her to be enrolled in this study. Some members of the team think that because the child is getting better without treatment, the risks of trial enrollment may outweigh the benefits. Others believe that the study is well designed, the patient is eligible, and the risks minimal. They call for an ethics consult.

This case highlights the complexities of navigating clinical trial enrollment for vulnerable populations. The patient cannot comprehend the trial and is therefore unable to assent to enrollment. Permission to enroll the child would have to be given by a representative of a state agency who will not be familiar with the child’s medical status or needs. This consenting agent thus relies on the patient’s medical team to objectively communicate the risks and benefits of trial enrollment.

The complexities of including vulnerable children in clinical trials are well known. This case emphasizes the added pressures placed on decision-making in times of uncertainty. COVID-19 is a novel and potentially fatal disease. It is suggested in global data that COVID-19 impacts children less frequently and less severely than adults; however, although rare, some children, and particularly those with medical complexity, experience morbidity and mortality related to this disease.1 

The clinical trial in this case is being used to evaluate a potentially beneficial treatment. The knowledge to be gained could be life saving for people with COVID-19. The child meets eligibility criteria. But is it in her interest to participate? How should this patient’s clinicians balance their obligations to facilitate research with their obligation to protect vulnerable children? There are obvious concerns surrounding trial enrollment of individuals who cannot assent to a drug with potential side effects, and yet the exclusion of vulnerable populations from clinical trials also carries heavy ethical implications. Early in the decision-making process, the care team should explore the existence or availability of a guardian ad litem, who may serve as an advocate for the patient separate from the court personnel likely to render a final enrollment decision.

The team must address 2 interrelated questions as they contemplate enrollment. First, they must consider whether this patient is an appropriate subject for this trial. She developed symptoms severe enough to warrant hospital admission and has a chest radiograph suggestive of pulmonary disease, thereby meeting technical inclusion criteria. However, the team must also consider whether there are valid reasons not to enroll her despite meeting these criteria. Some clinicians raised concerns about the study design. Specifically, they felt that children unlikely to receive benefit from the trial drug should be excluded. This patient’s study data would not likely contribute to answering the primary study question, yet enrollment could expose her to harm. They also worried that enrollment of many asymptomatic patients would hinder the assessment of drug efficacy. Others felt that known variability in pediatric presentations may inappropriately exclude children from important clinical research and the objective fulfillment of inclusion criteria was adequate support for trial enrollment.

Next, the care team must carefully consider the risks and benefits of enrollment for this individual patient and be prepared to explicitly and objectively discuss these issues with the child’s legal decision-maker. There are risks to all experimental medications. Those risks might be balanced by potential benefits if this child was critically ill. But she is not. Some clinicians felt that this patient was unlikely to receive any benefit from the trial drug given her stable clinical status. Others, acknowledging her medical complexity, felt that she was at higher risk for progression of COVID-19–related disease and complications than a healthy peer and that the trial drug had potential to improve the patient’s course of illness. The benefit to this patient is unclear, but will she be harmed? When considering trial enrollment, risks must be assessed and described with the same scrutiny as benefits, including potential side effects that may be unknown at the time of trial enrollment. The patient’s overall medical complexity and multidrug antiepileptic regimen both may increase her risk for adverse events associated with the trial drug.

Ultimately, the team decided that the patient’s clinical characteristics deemed her an appropriate trial subject.

In summary, the ethical issues surrounding trial enrollment for this patient are highly complex and the varying opinions expressed by this patient’s care team highlight the uncertainties posed by this case. A priori trial exclusion based on this child’s cognitive capacity is inappropriate because it would preclude her from the possible receipt of a potentially beneficial drug and may interfere with scientific advancement. The child meets the trial’s inclusion criteria. Balancing all factors in this case, it is reasonable to offer the option of trial enrollment. However, owing to the child’s clinical characteristics and the lack of an emotionally involved decision-maker, the onus is on this patient’s care team to provide a direct and balanced view of risks and benefits to the legal decision-maker, ideally with the involvement of an appointed advocate. The team must remain cognizant of biases and avoid the temptation to share subjective opinions. This may be particularly challenging in the setting of a pandemic; uncertainty is high, information sources are vast and variably reliable, and fear is prevalent. Despite these circumstances, it is critical to remember that enrollment in a clinical trial is never obligatory, and we must balance beneficence with nonmaleficence.

As a potential research subject, the girl described in this case manifests many characteristics of vulnerability when it comes to participation in research. As a person with neurocognitive impairment, she is part of a group whose members have experienced centuries of discrimination within and beyond the health care system. “Feeble-minded” people in the early 20th century were targets for eugenics laws that discouraged reproduction. Through the 1960s, people with disabilities in some states were sterilized without their consent. In the 1950s, researchers deliberately infected children who had intellectual disabilities with hepatitis B at the Willowbrook State School in New York.2  These historical examples of researchers and policy makers taking advantage of disempowered and vulnerable people should give us pause and lead us to be careful before enrolling a child like this in a research study.

Although such forms of outright discrimination are now illegal in the United States, implicit bias among physicians toward people with disabilities persists.3  As a person with neurocognitive impairment, the girl in this case is unlikely to have the capacity to understand and appreciate the consequences of participating in research. And even if she had the average cognitive ability of a 12-year-old child, she would still at most be able to assent to research participation. Parents of children with severe disabilities are often eager to try anything that might help their child, and the lure of the therapeutic misconception is likely higher among parents who may feel that they have much to gain. In this child’s case, she is a ward of the state, so it is not even clear who is designated to ensure that her best interests are at the forefront of any decision to participate in research.

Despite all these concerns, this child should be allowed to participate in research regarding treatment of COVID-19 or other research studies, assuming that the rules of human subject participation are scrupulously followed.4  Those rules would require the right answers to the following 6 questions. (1) Does the research have social or scientific value? (2) Will the study produce reliable results? (3) Is subject selection fair? (4) Are the risks minimal given that there is the prospect of benefit in a drug trial? (5) Is there proper oversight by an institutional review board (IRB)? (6) Is the permission of her legal guardian obtained after an ongoing process with transparency about the relevant risks and potential benefits of participation? If all of these conditions are met, then the child should be permitted to participate in the study. This is consistent with a core principle of the disability rights movement: people with disabilities should be fully included in all aspects of society, including the option of participating in research. Indeed, inclusion is essential if people with disabilities, including children in state custody, are to fully benefit from the fruits of scientific research.

Some points in this case deserve special attention. Unlike the Willowbrook case in the 1950s, this girl is not being asked to participate because of conditions related to her age or disability. She meets inclusion criteria for a study that is presumably being offered to every patient at the hospital who also meets the same criteria. This means that the selection of subjects is fair; she is not being singled out for participation on the basis of anything besides having COVID-19.

Another important safeguard of her rights would be to ensure that the IRB or the ethics consultation team included someone with a disability perspective. Although there is obviously a wide range of opinions among people with disabilities regarding research and consent more generally, just the presence of a person with a disability can help the entire group keep disability issues at the forefront.

Finally, we must address the thorny question of who can weigh the risks and benefits of research participation for this particular child given that she cannot and that she apparently does not have a legal guardian. States often appoint a guardian ad litem or other people who can represent the child’s best interest in addition to judicial oversight. The research and clinical teams are right to ask for a clinical ethics consult because the wisdom of a group of well-informed people, representing a wide variety of perspectives, can help a guardian or judge understand whether a reasonable person in this girl’s specific circumstances would want to join in the research study.

This case illustrates how challenging it is to manage competing interests in clinical research. Issues of equipoise must be considered in addition to the link between eligibility criteria and study power. When all these things are considered, I argue that the treating team has no duty to enroll this patient in this clinical trial. Doing so could even be misguided.

Clinical investigators have many motivations to enroll patients in clinical trials. They want to learn more, publish more, stay on budget, and get additional funding to do more studies. In balancing these motivations, clinical investigators must prioritize first the interests of the patient even if doing so threatens those core academic livelihoods. For example, clinical investigators must conduct complete and noncoercive informed consent conversations with potential subjects and disclose any financial conflicts of interest.

Not all competing interests arise from clinical investigator self-interest.5  Altruistic desires to help colleagues succeed or to uncover new scientific knowledge are powerful, nonselfish motivators.

To manage competing interests, clinical investigators should examine which interest is most important given their role in the moment. If the clinical investigator is seeing a patient in the throes of an acute illness, protecting the patient’s clinical stability is the most compelling interest. If enrolling in a study threatens that, the clinical investigator might appropriately decide that receipt of an investigational agent poses unacceptable risks.

Here, the clinicians taking care of a 12-year-old girl amid a life-threatening exacerbation of chronic epilepsy are asking if they have an obligation to enroll her in a clinical trial of a new drug against COVID-19. Their first priority should be to consider whether study enrollment could endanger their patient’s full and continued recovery from seizures. After all, we do not know if the drug could have unknown neurologic side effects or unknown propensity to interact with antiseizure medications.

Clinician and investigator prioritization of the clinical interests of potential research subjects can be further challenged by therapeutic misconception, namely, the belief that experimental therapy is likely beneficial when, instead, the drug is as likely to be harmful as beneficial.

To guard against this temptation, all should remember the concept of equipoise. Equipoise is “a state of genuine uncertainty on the part of the clinical investigator regarding the comparative therapeutic merits of each arm in a trial.”6 

The existence of equipoise, which is required for trial conduct to be ethical, explains why there is no duty to enroll any patient in any specific study: there is no duty to roll the dice with an experimental therapy that could help, hurt, or do nothing.

Even if the patient’s clinical interest in recovering from seizures were not potentially endangered by receipt of an experimental antiviral drug and even if her clinicians are disabused of therapeutic misconception, enrolling this patient into the proposed study could undermine study power. A study that is not well designed and that therefore cannot answer the study question is, by definition, unethical. The statistical power of the study must be adequate to answer the study’s core research question. Investigators must enroll enough subjects, for instance, and measure end points that are reliable enough to make it possible to detect a clinically meaningful result.

Subject inclusion and exclusion criteria contribute to study power by ensuring subjects are reasonably likely to exhibit a meaningful response to therapy. The inclusion of subjects with no signs of illness and thus who are incapable of improving in response to an experimental intervention will undermine study power. The inclusion of subjects who are so near death that they will deteriorate despite intervention can also undermine study power in much the same way. Excess inclusion of subjects of either type could lead to erroneous conclusion that the study drug was ineffective.

In this situation, the patient was screened for COVID-19 presumably because she had a fever. If her fever resulted from her seizure and was unrelated to COVID-19,7  the patient cannot improve because of experimental antiviral therapy and her inclusion in the study will reduce study power. Alternatively, if her fever did result from COVID-19, she might (or might not) respond to the experimental antiviral, but her intercurrent seizure of uncertain relationship to COVID-19 could undermine the generalizability of either finding to other patients with more common manifestations of COVID-19. Thus, the study’s inclusion and/or exclusion criteria should preclude the inclusion of patients like this.

Because neither the patient’s clinical safety nor the clinical trial’s validity are likely to be served by enrolling her, the clinical team should abandon the notion they have an obligation to enroll her in the study. In addition, study investigators and/or her clinicians should weigh whether it would be scientifically disadvantageous to enroll her in this study.

This is clearly a case in which reasonable people may disagree about whether it is ethically appropriate to enroll this patient in this study. The study was, presumably, approved by relevant IRBs. That does not, however, absolve clinicians of the responsibility to decide whether the study is appropriate for their own patient. In this case, different doctors came to different conclusions about that. That is how it should be. Some such patients, then, will be enrolled and some will not. That is okay, too. No study is perfect. But with careful consideration of the relevant science and ethics, studies should be ethically defensible. This study is ethically defensible, as is enrolling a patient like this. That does not mean that enrolling such a patient is ethically obligatory.

Drs House and Shubkin conceptualized and designed the article and drafted the initial manuscript; Drs Lahey and Brosco contributed to the initial manuscript draft; Dr Lantos advised on the design and content of the article; and all authors reviewed and revised the manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.

FUNDING: No external funding.

COVID-19

coronavirus disease 2019

IRB

institutional review board

1
Shekerdemian
LS
,
Mahmood
NR
,
Wolfe
KK
, et al
;
International COVID-19 PICU Collaborative
.
Characteristics and outcomes of children with coronavirus disease 2019 (COVID-19) infection admitted to US and Canadian pediatric intensive care units [published online ahead of print May 11, 2020]
.
JAMA Pediatr
.
doi:10.1001/jamapediatrics.2020.1948
2
Krugman
S
,
Ward
R
.
Clinical and experimental studies of infectious hepatitis
.
Pediatrics
.
1958
;
22
(
5
):
1016
1022
3
Wilson
MC
,
Scior
K
.
Attitudes towards individuals with disabilities as measured by the implicit association test: a literature review
.
Res Dev Disabil
.
2014
;
35
(
2
):
294
321
4
Feudtner
C
,
Brosco
JP
.
Do people with intellectual disability require special human subjects research protections? The interplay of history, ethics, and policy
.
Dev Disabil Res Rev
.
2011
;
17
(
1
):
52
56
5
Bero
L
.
Addressing bias and conflict of interest among biomedical researchers
.
JAMA
.
2017
;
317
(
17
):
1723
1724
6
Freedman
B
.
Equipoise and the ethics of clinical research
.
N Engl J Med
.
1987
;
317
(
3
):
141
145
7
Arons
MM
,
Hatfield
KM
,
Reddy
SC
, et al;
Public Health–Seattle and King County and CDC COVID-19 Investigation Team
.
Presymptomatic SARS-CoV-2 infections and transmission in a skilled nursing facility
.
N Engl J Med
.
2020
;
382
(
22
):
2081
2090

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.