Article Text

Download PDFPDF

Mistrust and inconsistency during COVID-19: considerations for resource allocation guidelines that prioritise healthcare workers
  1. Alexander T M Cheung,
  2. Brendan Parent
  1. Division of Medical Ethics, New York University School of Medicine, New York, New York, USA
  1. Correspondence to Alexander T M Cheung, Division of Medical Ethics, New York University School of Medicine, New York, New York, USA; alexander.cheung{at}


As the USA contends with another surge in COVID-19 cases, hospitals may soon need to answer the unresolved question of who lives and dies when ventilator demand exceeds supply. Although most triage policies in the USA have seemingly converged on the use of clinical need and benefit as primary criteria for prioritisation, significant differences exist between institutions in how to assign priority to patients with identical medical prognoses: the so-called ‘tie-breaker’ situations. In particular, one’s status as a frontline healthcare worker (HCW) has been a proposed criterion for prioritisation in the event of a tie. This article outlines two major grounds for reconsidering HCW prioritisation. The first recognises trust as an indispensable element of clinical care and mistrust as a hindrance to any public health strategy against the virus, thus raising concerns about the outward appearance of favouritism. The second considers the ways in which proponents of HCW prioritisation deviate from the very ‘ethics frameworks’ that often preface triage policies and serve to guide resource allocation—a rhetorical strategy that may undermine the very ethical foundations on which triage policies stand. By appealing to trust and consistency, we re-examine existing arguments in favour of HCW prioritisation and provide a more tenable justification for adjudicating on tie-breaker events during crisis standards of care.

  • resource allocation
  • allocation of health care resources
  • public health ethics
  • policy guidelines/inst. review boards/review cttes

This article is made freely available for use in accordance with BMJ’s website terms and conditions for the duration of the covid-19 pandemic or until otherwise determined by BMJ. You may use, download and print the article for any lawful, non-commercial purpose (including text and data mining) provided that all copyright notices and trade marks are retained.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Soaring demand for healthcare resources dedicated to patients with COVID-19 early in the pandemic led many bioethicists to discuss what to do if ‘the toughest triage’ ever occurred: namely, the allocation of ventilators if the needs of patients exceeded a hospital’s capacity to provide care.1 Thanks to monumental efforts made by healthcare and other essential workers, hospitals in the USA narrowly avoided having to implement such triage policies (‘crisis’ standards of care). However, as many states contend with the second surge in cases and a faltering commitment to social distancing measures, institutions may soon need to answer the unresolved question of who lives and dies when ventilator demand exceeds supply.

Although most US triage policies have seemingly converged on a commitment to saving as many lives as possible using clinical need and benefit as primary criteria for prioritisation,2 significant differences exist between institutions in how to assign priority to patients with identical prognoses: the so-called ‘tie-breaker’ situations. In particular, one’s status as a frontline healthcare worker (HCW) has been a proposed criterion for prioritisation in the event of a tie.3–5 Some have commented that this addendum to standard triage protocol, while seemingly intuitive, is a mistake.6

In the following, we outline two major grounds for reconsidering HCW prioritisation. The first recognises trust as an indispensable element of clinical care and warns that perceptions of favouritism could exacerbate an already pervasive mistrust of the healthcare system. The second concerns the ways in which proponents of HCW prioritisation use justifications that deviate from the very ‘ethics frameworks’ that serve to guide decision-making during resource allocation—a rhetorical strategy that may undermine the very ethical foundations on which triage policies stand. Through lenses of trust and consistency, this article seeks to both rebut existing arguments made in favour of HCW prioritisation and provide a more tenable justification for adjudicating on tie-breaker events. The methodology proposed herein may also have implications for the allocation of other scarce treatments and preventive measures such as vaccines.

A trust-oriented response to resource allocation

Long before ‘COVID-19’ became a household phrase, several bioethicists had noted a very different epidemic spreading throughout the healthcare system in China. This emerging public health threat was not confined to Chinese hospitals, either: the ‘epidemic of mistrust’ had indeed become a global crisis.7 A rise in antivaccination campaigns, violence against physicians and end-of-life controversies in recent years all serve as chilling reminders that mistrust has come to characterise public perceptions of healthcare.7–9 If a trend of mistrust was concerning prior to COVID-19, it is now a recalcitrant obstacle to any effective public health response to the virus. Recent polls show less than half of US citizens plan to get a coronavirus vaccine when it becomes available, and misinformation—or downright conspiracy theories—abounds regarding mask usage.10 11 How should the concept of trust, then, influence triage policies for scarce resource allocation?

Trust in routine and crisis standards of care

Trust makes several cameos in the various US policies available. One policy admits that ‘public trust will be essential to ensure cooperation with restrictive public health measures’ and offers some considerations to mitigate potential mistrust due to ‘perceptions of unfairness’.12 The New York State (NYS) Ventilator Allocation Guidelines also remark that trust must be established through public deliberation and transparency while cautioning that certain distribution schemas ‘may penalize individuals who are […] distrustful of the health care system’.13 Notably, these individuals more often than not tend to divide along racial or socioeconomic lines.9 Guidelines have thus presented trust as instrumental in realising the public health agenda and distrust as inimical to achieving a just allocation of ventilators. These consequentialist views of trust, unfortunately, do not explicate with much detail how mistrust has become so rampant. Nor do they offer much in the way of substantive ethical guidance through a lens of trust besides the general appeal to transparency and deliberation (two topics we shall return to later). This is not wholly surprising. As Nie et al write: ‘like air to life, trust is too often taken for granted, and its vital role is rarely acknowledged and valued unless it is threatened or lost’.7

Nie and colleagues, admittedly, claimed a crisis of interpersonal patient–physician trust—not trust in a public health response. Indeed, it is often said that crisis standards deviate from routine clinical care out of practical necessity and divergent ethical requirements.1 13 Some have gone as far to say that a pandemic is itself exceptional when compared with ‘typical’ mass casualty emergencies such as earthquakes due to the longevity and pervasiveness of the crisis.4 The primary difference between crisis and routine standards, generally speaking, is a reduced emphasis on the individual needs of a patient to instead prioritise the aims of maximal rescue and distributive justice as a consequence of resource scarcity.12

But if trust genuinely constitutes an enduring and fundamental aspect of routine medical care,9 should it be thrown out with the bathwater during the pandemic? Some commentators on ventilator triage think not. By establishing independent triage committees, these commentators hope to preserve the import of trust from a deontological perspective. Triage committees would therefore allow physicians and nurses ‘to maintain their traditional roles as fiduciary advocates’ for patients while protecting them from the moral distress of triage deliberation.1 The implicit assumption here is that medicine without trust is not a medicine at all.

HCW prioritisation may exacerbate mistrust

At the outset of the pandemic, concerns arose about an asymmetric distribution of healthcare resources. News reports suggested that politicians, celebrities and the rich were receiving preferential access to testing despite shortages in communities, leading some forsaken patients to ask, ‘Why are they getting in front of the line? People like me, average Joes, we get pushed to the back of the line’.14 15 Yet perceptions of biased rationing in healthcare have long preceded the pandemic; in the 1990s, a public outcry ensued after select high-profile individuals in the USA received organ transplants seemingly (and in one case literally) overnight.16 17

The WHO cautions that any triage policy ‘should be developed with great care, given the danger that those which favour certain categories of workers may be perceived as unfair and undermine public trust’.18 What would happen to public trust if ventilators are preferentially allocated away from patients, and instead given to the very individuals entrusted to treat those patients? Empirical data are lacking on this question and prior community-based studies have proven equivocal. Some have suggested at least public support of HCW prioritisation for preventive measures like vaccines,19 while one study found that prioritising individuals based on utility proved ‘complex and difficult to operationalize’ due to varying interpretations of one’s value.20

What these studies have invariably recommended, however, is that any approach to resource allocation ought to incorporate public opinion to foster trust and compliance with public health measures—‘both of which are more likely if the development process has been transparent and inclusive’.21 Indeed, community-based deliberation has proven invaluable in research ethics to improve institutional accountability and public trust, especially among traditionally marginalised groups.22 The latter point is particularly salient for the current pandemic and its disproportionate impact on minority communities.15

For example, African Americans often report greater mistrust of the healthcare system due to historical experiences of abuse and suboptimal care. Although medical mistrust contributes to lower rates of health-seeking behaviours and preventive measures, it conversely correlates with higher rates of emergency department utilisation.23 In the case of COVID-19, the irony is that those most likely to mistrust the healthcare system are those most likely to require intensive care resources. By denying this population access to ventilators—in favour of a healthcare provider, no less—the effect could exacerbate existing mistrust.

Ventilator allocation policies are already somewhat of an ‘inside job’ when it comes to deliberation. They are usually written in part by HCWs themselves, and triage committees most often consist of physicians or nurses.2 24 Given the controversial nature of HCW prioritisation, even among those privy to the debate, it seems that the outward appearance of favouritism is inevitable without substantial deliberation including input from the public. Through a lens of trust, we argue there is a strong a priori argument to avoid the preferential prioritisation of any one class of workers over another—barring convincing evidence that suggests otherwise.

Trustworthiness through consistency: the ‘ethics framework’ in triage policies

Although proponents of HCW prioritisation have largely ignored the import of trust thus far, they have instead offered supplemental arguments or principles such as reciprocity and narrow social utility to defend their position.3–5 Rather than highlighting the intrinsic problems of these arguments,6 17 25 however, it is worth pointing out that these justifications rarely (if ever) appear in the corresponding ‘ethics frameworks’ that existing triage policies espouse. This inconsistency is ethically concerning, particularly through a lens of trust and trustworthiness.

From the public perspective, empirical data have demonstrated that consistency is one of the most critical components of trust in health messaging and resource allocation.19 26 As one participant in a recent study comments, ‘Overall there is nothing fair about any of this [allocating scarce resources in an emergency]. So, the only way you can kind of say you’re being fair is to be consistent’.27 From the institutional perspective, however, consistent application of ethical principles is a procedural requisite to being trustworthy rather than merely earning trust 28—a subtle distinction that has received much attention in recent years.29

The following section thus explores the inconsistencies that arise when prioritising HCWs for scarce resources and the implications for public trust during the pandemic. It also asks whether existing ethics frameworks are intrinsically insufficient to handle tie-breaker events—thereby requiring extraneous means of adjudication—or if ad hoc amendments like ‘reciprocity’ actually undermine the trustworthiness and ethical groundings that triage policies purport to stand on.

The landscape of US ventilator allocation policies

Policies for resource allocation in a public health emergency must tread carefully between two conflicting requirements. On the one hand, they must provide sufficient structure and substance to provide clear guidance for clinicians during a disaster to alleviate the burden of decision making and provide legal protection. On the other hand, policies need certain flexibility and responsiveness to allow proportional and appropriate responses to a given circumstance.1 13 Ethical frameworks therefore often preface policies in order to provide principles, goals or values to both guide and defend decisions made within the policy itself, so long as the policy is only enacted in a defined set of circumstances (eg, a worldwide pandemic under true intensive care unit resource scarcity).

Although the NYS Ventilator Allocation Guidelines asserts, ‘An ethical framework must serve as the starting basis for a plan that proposes to allocate ventilators fairly’, scholarship on public policy has shown such frameworks may not provide as firm of a foundation for action as intended. Indeed, one group investigated dozens of public policies with espoused ‘ethics frameworks’ to determine if they function as true foundations, scaffolds, or decorative ‘frames’ that merely serve to legitimise policies through the imprimatur of ethical deliberation.30 Their results showed that different frameworks often diverge in substance, language and even interpretations of identical terms.

A review of ventilator triage policies from US hospitals found that the five most common ethical principles mentioned in such policies were: justice; transparency; stewardship (duty to steward resources); duty to care; and duty to prevent unnecessary loss of life.2 Of the 26 unique policies reviewed, 10 gave preference to HCWs: 5 on the basis of reciprocity; 4 on narrow social utility; 1 on both; and 2 on no substantive basis whatsoever. Definitionally, reciprocity refers to a societal duty to provide HCWs with certain privileges in recognition that they have assumed great risk in serving the greater good.15 Narrow social utility refers to the instrumental value of an individual that is relevant and specific to the public health emergency at hand.4

It is surprising that the analysed triage policies do not ground or consider HCW prioritisation in terms of the five most common framing ethical principles and occasionally ignore some altogether. Is the ethics framework indeed just a decorative frame, as Giacomini and colleagues put forward,30 that fails to do the required normative or formative work? Or were the conclusions to which these ethical principles led simply deemed unsatisfactory, thus requiring additional ethical legitimisation for HCW prioritisation?

A lack of transparency

As the NYS Ventilator Allocation Guidelines remark, ‘[a] plan that does not directly incorporate ethical considerations into its clinical protocol is unlikely to withstand ethical scrutiny’.13 To add to this statement, any plan that does not consistently adhere to its ethical considerations is unlikely to withstand ethical scrutiny as well. Let us examine the principle of transparency first, which is often touted as a key to both facilitating procedural justice and engendering trust.2 3 Ironically, many of the policies that Antommaria and colleagues reviewed were shared confidentially and were not available for examination. Another review of US ventilator allocation guidelines found that only 26 states had publicly available documents describing how allocation would or should occur.24 If transparency is indeed a prerequisite to trust, then it seems many institutions have inconsistently followed their own ethical requirements—and thereby undermined their own trustworthiness.

We concede that the public may not always need or want to know ‘how the sausage is made’. A delicate balance has historically existed between the care a patient actually receives and the ‘back office’ accounting decisions that distribute finite resources.31 Similarly, triage policies inherently consider matters of life and death—topics that HCWs and bioethicists routinely discuss—and the lay response may be a knee-jerk reaction. The media attention given to a leaked letter concerning ventilator allocation guidelines in one hospital system suggests how provocative the topic can be.32

But then the question arises: why emphasise transparency as a guiding principle at all? If transparency is as important (and necessary for trust) as the frameworks suggest, a commensurate explanation for a policy’s confidentiality ought to be provided as well. This inconsistency is unacceptable. As mentioned previously, deliberative democracy methods can foster public trust and identify salient or contentious features of an allocation policy. But if the condition of transparency has not been met, then a schema in which HCWs write and enact guidelines that also prioritise HCWs carries the risk of engendering mistrust.

The duty to care versus reciprocity

Let us next explore the tension between the duty to care and the concept of reciprocity as it pertains to HCW prioritisation. Reciprocity argues that services rendered to society at significant risk ought to be repaid, in this case, with preferential access to ventilators in the case of a tie.15 33 Some have also suggested that the promise of reciprocity alone may discourage absenteeism.3 We argue that appealing to reciprocity inherently undermines the ethical import of the duty to care (and vice versa) and may therefore demonstrate another inconsistency with ethics frameworks espoused by many allocation guidelines.

Specifically, the duty to care (sometimes ‘duty of care’) as defined in several allocation guidelines seems to temper some of the claims one might make in the name of reciprocity. For example, one remarks that the ‘duty to care is the fundamental obligation for providers to care for patients’ and ‘physicians must not abandon, and patients should not fear abandonment’ by their care providers.13 Another notes that the duty to care imposes on HCWs ‘a duty to expose themselves to some, although not unlimited, risks’.34

Insofar as the duty to care is, in fact, a duty tied with corresponding obligations, it is far from obvious that reciprocity vis-à-vis ventilators is owed to HCWs. As the latter quote earlier suggests, some risk is part and parcel of most jobs in healthcare. Reciprocity therefore runs counterintuitive to the duty to care: if a guiding principle of ventilator triage policies (1) implies some risk is inevitable during a pandemic and (2) prohibits abject abandonment of patients, appealing to reciprocity as a response to risk and absenteeism undermines the obligatory nature of that duty. There are also practical considerations when using ventilators as the currency of reciprocity; any unit allocated to an HCW under true scarcity must inevitably come from another patient. The spatiotemporal overlap of the duty to care and duty of reciprocity, at least in regards to ventilators, might inherently place them at odds.

We agree that some actions under certain conditions may be supererogatory—that is, beyond the call of the duty to care. We also recognise that some forms of reciprocity may have stronger claims, so long as they do not interfere with the duty to care. Indeed, some have demanded reciprocal obligations of hospitals and governments to protect HCWs through adequate infection control measures, all while cautioning that reciprocity through ventilator allocation—which makes demands of patients as well as institutions—‘could be influenced by or perceived to be influenced by self-interest … [and] requires substantial community support’.34

Given that the duty to care is espoused in the ethics framework of many allocation guidelines, we argue that it ought to receive lexical priority over reciprocity. Although the desire to thank HCWs for their courageous efforts is both intuitive and encouraged, we believe the preferential prioritisation of ventilators for this group is inconsistent with the duty to care and has the inherent risk of exacerbating mistrust.

Stewardship or social utility?

Some proponents of HCW prioritisation have used the concept of ‘narrow social utility’ to make judgements about the instrumental value of certain jobs or positions.3 12 They argue that their narrowed definition of utility differs from biased evaluations made about one’s general societal worth because the outcomes under consideration are short term and specific to the pandemic.4 Following this line of reasoning, HCWs ought to receive prioritisation for ventilators because they are, or will be, instrumental in realising the goals of triage.

Although not explicitly found in any available ethics framework, the narrow social utility of HCWs is admittedly pertinent to the success of crisis standards of care. But several authors have noted the practical difficulties that arise when attempting to establish which frontline HCWs are most deserving of prioritisation—a process that would require either ‘Solomon-like determinations’17 or ‘subjective judgments of utility and risk’ at best.25 Even proponents of HCW prioritisation have equivocated on the issue of exactly who gets priority by acknowledging that the topic remains ‘fraught with problems’.33

We echo that narrow social utility is a procedurally fickle principle to operationalise and therefore prone to inconsistencies.13 20 Instead, we propose a principle found in the ethics framework of most triage policies: stewardship.2 This considers ‘the obligation for government and healthcare providers to responsibly manage resources during a period of true scarcity’.13 Largely construed as prudent rationing of services or supplies, stewardship is a requisite for saving as many lives as possible throughout the duration of a pandemic. Ferguson and Caplan come closest to defending HCW prioritisation using this line of reasoning with their concept of sustainability which, broadly speaking, is ‘a matter of securing the long-term stability, capacities, and integrity of institutions and the medical profession’.35

Although perhaps still controversial in its application, stewardship is itself a much more tenable argument than ‘narrow social utility’ when prioritising HCWs in a tie-breaker situation. For one, the principle of stewardship can already be explicitly found within the majority of ethics frameworks.2 It also lends itself well to resource allocation through a lens of trust. By avoiding the subjectivity and inconsistency inherent in applying narrow social utility, one can mitigate potential perceptions of favouritism. Likewise, insofar as mistrust erodes the foundations of medicine, any action that damages trust’s integrity and stability ought to be considered ‘unsustainable’—and therefore ethically unacceptable.

Stewardship also circumvents some of the practical issues and appearances of favouritism left unresolved by a narrow social utility. Although no human is ever ‘fungible and replaceable’,4 some jobs, in reality, are. But stewardship—unlike narrow social utility—does not require a premature judgement of one’s value at the outset of the pandemic. It may prioritise flexibly and appropriately as particular workers become scarce. Notably, this may include those whose roles are entirely integral to hospital function but may not necessarily include physicians or nurses. Stewardship of trust also demands that any decision to prioritise certain groups ought to be transparent and engage the public as a rule. A duty to steward human resources recognises both HCWs and public trust as indispensable means to end the pandemic while simultaneously obligating institutions and governments to ensure their continued well-being.


To reiterate an earlier point, triage policies for resource allocation must provide both adequately flexible yet sufficiently concrete guidance during a pandemic. This article aims to strike that balance by admittedly not taking a strong stance on the ethical permissibility of HCW prioritisation per se. Our chief concern arises from a long history of mistrust in healthcare that has proven a hindrance to any effective public health strategy in the USA. Exacerbating the appearance of favouritism alone should give us pause before redirecting life-saving resources to those most often in charge of writing and implementing triage policies.

Bioethics should also hold itself to standards of similar rigour to medical research. We demand that scientists assess novel vaccines or repurposed antiviral medications thoroughly before subjecting the public to their costs and effects. So too should bioethicists commit to the set of overarching principles—or ethics frameworks—that preface so many allocation guidelines. Trustworthiness in medicine requires adherence to fundamental ethical principles.28 Why should tie-breaker situations be exempt? Appealing to additional ethical principles without explaining why the original ethics framework is inadequate is at least ethically inconsistent and at worst unethical. Consistency, like transparency and community-based deliberation, can preserve trust. The public response has suggested as much: inconsistent messaging regarding mask usage and asymptomatic spread severely marred the credibility of public health officials and institutions in the USA.

We suspect, however, that the espoused principles can sufficiently guide resource allocation—even beyond ventilators. If prioritising HCWs for preventive measures like vaccines or personal protective equipment help enable the ‘duty to care’, comport with a ‘duty to steward resources’, and ‘prevent unnecessary loss of life’, then extraneous or supplanting principles are clearly unnecessary. Likewise, if the original principles cannot justify HCW prioritisation for ventilators under most circumstances, then either principle like reciprocity or narrow social utility should have been included in the original framework—or HCWs should not be prioritised at all. In order to be useful in tie-breaker situations, triage policies must inevitably delineate whom to prioritise. But if those decisions are opaque, inconsistent or disproportionately exclude a vast number of patients, then we may draw a line that leaves a mistrustful public on the other side.



  • Twitter @parent_brendan

  • Contributors AC developed the concept for the article with consultation and expansion from BP. AC performed the literature review and primary drafting, while BP provided revisions and expansion on concepts. AC is the guarantor for the overall content.

  • Funding AC and BP acknowledge support from the NYU Rudin Medical Ethics and Humanities Fellowship (No Grant/Award Number).

  • Competing interests None declared.

  • Patient consent for publication Not required.

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

  • Data availability statement There are no data in this work.