Les principes d’équité et d’utilité dans l’allocation des ressources limitées en situation de pandémie Principles of equity and utility in allocating limited resources in pandemics

Résumé Abstract La pandémie de COVID-19 remet à l’honneur la question éthique de l’allocation des ressources limitées, en termes d’accès à des soins intensifs et à des respirateurs. Se pose la question éthique suivante : sur quels principes éthiques se baser pour effectuer le triage des patients qui auront accès aux ressources quand elles sont insuffisantes pour répondre aux besoins de tous? Pour en débattre, deux références historiques de triage sont d’abord présentées ; l’une s’appuie sur un principe égalitaire de réponse aux besoins individuels, l’autre sur un principe d’utilité sociale. Après avoir défini les conditions d’équité en tant qu’égalité procédurale et réponse adéquate aux besoins, deux types de protocoles sont étudiés en mettant l’accent sur les critères d’équité et d’utilité qu’ils préconisent. Les types de protocoles sont ensuite comparés en présentant leurs forces et leurs limites dans la réponse qu’ils apportent aux besoins populationnels et individuels. Notre analyse met en évidence la difficile conciliation entre les objectifs populationnels et les objectifs cliniques en situation de pandémie, tout en montrant qu’un protocole qui utilise comme outil le Sequential Organ Failure Assessment (SOFA) facilite cette conciliation. The COVID-19 pandemic brings to the forefront the ethical question of allocating scarce resources in terms of access to intensive care and ventilators. The ethical question is: on what ethical principles should we base the triage of patients who will have access to resources when they are insufficient to meet the needs of all? In order to discuss this issue, two historical references for triage are first presented; one is based on an egalitarian principle of meeting individual needs, the other on a principle of social utility. After defining the conditions of equity as procedural equality and adequate response to needs, two types of protocols are examined with a focus on the criteria of equity and utility they advocate. The types of protocols are then compared by presenting their strengths and limitations in responding to population and individual needs. Our analysis highlights the difficulty of reconciling population and clinical objectives in a pandemic situation, while showing that a protocol that uses the Sequential Organ Failure Assessment (SOFA) tool facilitates this reconciliation.

The COVID-19 pandemic brings to the forefront the ethical question of allocating scarce resources in terms of access to intensive care and ventilators. The ethical question is: on what ethical principles should we base the triage of patients who will have access to resources when they are insufficient to meet the needs of all? In order to discuss this issue, two historical references for triage are first presented; one is based on an egalitarian principle of meeting individual needs, the other on a principle of social utility. After defining the conditions of equity as procedural equality and adequate response to needs, two types of protocols are examined with a focus on the criteria of equity and utility they advocate. The types of protocols are then compared by presenting their strengths and limitations in responding to population and individual needs. Our analysis highlights the difficulty of reconciling population and clinical objectives in a pandemic situation, while showing that a protocol that uses the Sequential Organ Failure Assessment (SOFA) tool facilitates this reconciliation.
Les critiques les plus sévères envers ce que Shana Alexander (6) a nommé le God Committee 3 sont venues de la part de philosophes et d'éthiciens partisans de l'égalité dans la répartition des ressources en santé à l'encontre des approches utilitaristes, qui favorisent le plus grand bien pour le plus grand nombre, au lieu de mettre l'accent sur les droits et les besoins individuels. Du point de vue utilitariste, pour favoriser le plus grand bien pour le plus grand nombre, l'action ou la règle morale à choisir doit être celle qui apporte le plus de bénéfices et les moindres torts à toutes les personnes qui en seront affectées. Il s'agit là d'une théorie morale qui se base sur les conséquences de l'action ou de la règle. Si l'action ou la règle visée apporte plus de bénéfices que de torts au plus grand nombre de personnes, voire à la population en général en période de pandémie, elle est moralement bonne.

Principles of equity and utility in allocating limited resources in pandemics
The COVID-19 pandemic brings to the fore the ethical issue of allocating limited resources, in terms of access to intensive care and ventilators. This context raises important ethical questions since some people will not survive because they will lack access to these resources, despite meeting the usual inclusion criteria for access to such services. The ethical question is the following: what ethical principles should be used to decide who should have priority for ICU intervention when resources are insufficient to meet everyone's needs? Should equity, which is what is normally applied in clinical practice, have priority in access to care? Alternatively, should utility, which is the greatest good for the greatest number of people, and generally pertains to the domain of public health, be prioritized?
To discuss this issue, two historical triage reference cases, one based on an egalitarian principle of responding to needs, the other on a principle of social utility, are first presented. After having defined the conditions of equity as procedural equality and adequate response to needs, two types of protocols are studied, incorporating criteria related to equity or utility. We then compare the types of protocols to highlight their contribution and limitations in meeting population and individual needs. Finally, our study concludes with some recommendations, notwithstanding the difficulty of reconciling population and clinical objectives in a pandemic situation.

SOME HISTORICAL REFERENCES CONCERNING TRIAGE
Physicians have long been concerned with the criteria to determine who will have priority access to care and treatment among war wounded, victims of cataclysms or acts of terrorism, and victims of pandemics. The pandemic context has often been compared to that of the battlefield and, from the outset, triage 15 has been associated with a utilitarian approach: the injured who received priority care are those who are least injured and who could return to the battlefield once they are healed. This was not the approach of Dr. Jean Dominique de Larrey, surgeon-in-chief of Napoleon's armies, who proposed a revolution in the Ancien Régime's ways of doing things by treating the most injured first, instead of giving priority to those who were more highly ranked (1,2). In so doing, he applied the egalitarian ideals of the French Revolution. This corresponds to the triage carried out today in emergency rooms and hospital intensive care units. It differs both from allocation according to the first come, first served principle, which corresponds to the waiting list in hospitals, or allocation by means of age and rank, which are social criteria, in contrast to the medical criterion of giving priority to the most affected, the ones in greatest need, in the sense that, without treatment, death follows.
In a situation of scarce resources, utilitarian criteria often take precedence; utilitarianism aims at the greatest good for the greatest number by favouring those who are most useful to society. When renal dialysis was opened to all patients with uremia, in the early 1960s, the Seattle Artificial Kidney Center had only nine machines. Questions were then raised about the criteria for choosing the patients who will have access to dialysis treatment. Two committees were created, and the evaluation of candidates was done in two stages: 1) medical evaluation: were excluded persons with diabetes, hypertension, pulmonary artery disease, and cerebrovascular disease; the committee automatically eliminated persons under 18 years of age and over 45 years of age; 16 2) psychosocial and economic evaluation based on age, gender, marital status, number of dependents, income, training, employment and past achievements (3).
The first assessment was conducted by nephrologists, and the second by a committee of members of the public (a housewife, a businessman, and a worker's representative), a lawyer and two physicians who were not nephrologists. This committee made a selection by prioritizing those who were most useful to society: the father of the family over the bachelor, the worker over the unemployed or the criminal, the resident of Washington State over the resident from another state. Criticism came from all sides, particularly because, consciously or unconsciously, the members of this committee had chosen people who resembled themselves (4,5). In any case, only the better-off had access to dialysis at that time, since it cost $10,000 per year for treatment. 17 The harshest criticisms of what Shana Alexander (6) called the God Committee 18 have come from philosophers and ethicists who advocated equality in the distribution of health resources over utilitarian criteria, which favour the greatest good for the greatest number, instead of focusing on individual rights and needs. From a utilitarian perspective, to promote the greatest good for the greatest number, the action or moral rule chosen ought to be the one that brings the greatest benefit and the least harm to all those who will be affected. Utilitarianism as a moral theory is based on the consequences of the action or rule. If the action or rule chosen brings more benefits than harm to the greatest number of people, or even the general population in a pandemic, it is morally good.
Utilitarian calculations are not absent from ethical analyses that propose triage protocols. Emanuel et al (7,8) present a utilitarian version of triage that is used in current protocols during the COVID-19 pandemic. They interpret the utilitarian principle as saving the most lives by maximizing the number of years to live (7). This interpretation is contrary to the biomedical ethical principle of beneficence, which advocates first treating the sickest, i.e., those who are more at risk of dying, regardless of social or demographic considerations, a principle that Dr. de Larrey applied to the triage of war wounded and that is applied in normal times in emergency departments and in access to intensive care. According to Emanuel et al., it would be legitimate during a pandemic to maximize the number of patients "that survive treatment with a reasonable 19 life expectancy and to regard maximizing improvements in length of life as a subordinate aim" (7, p.4); the latter criterion applies only if patients have an equivalent survival prognosis. From this point of view, the life cycle is used as a triage criterion. This principle is based on the idea that "each person should have an opportunity to live through all the stages of life (…) -to be a child, a young adult and to develop a career and family and to grow old -and to enjoy a wide range of opportunities during each stage" (8, p.854-855). The explanation given is presented as an egalitarian individual right to live one's life, but in fact, this practice consists of favouring the youngest at the expense of the oldest; it is age that is used as a social criterion of exclusion and, therefore, as a discriminatory criterion, as defined by the Canadian and Quebec Charters of Human Rights and Freedoms (9,10) and the Universal Declaration of Human Rights (11). In addition, there is a criterion based on the instrumental value of the person, since for an equivalent prognosis, it consists of treating caregivers before others, for the reason that they are necessary for the treatment of infected persons (7). These criteria are utilitarian in nature because they favour those who are most useful to society.

UTILITARIAN AND EGALITARIAN THEORIES OF JUSTICE
For utilitarian theories (12,13), individual rights and needs only count in the balance if they coincide with the general interest which, in the case of these theories, corresponds to the sum of individual interests. Applied to the field of health resource allocation, the utilitarian approach focuses on the probability of survival of those who will be able to benefit from limited resources. In other words, it is not the need, as assessed by the physician, that counts, but rather the prognosis of survival and the number of years to live, according to statistical analysis. From this point of view, it is generally the youngest and least afflicted who have the best prognosis, so they would be given priority in the allocation of limited resources. While it could be ethically justified to apply the principle of utility in public health because of its population-based focus, it is unethical to apply this principle in the clinical assessment of individual needs.
Advocating prioritizing the needs of those most affected rather than those most useful to society implies a medical assessment of the individual's health, which is not possible with a choice made on the basis of age, rank, or any other social criterion, including the criterion of utility. In fact, equity in egalitarian terms is defined according to two conditions: 1) the application of the same rule for all those who qualify, and 2) an adequate response to health needs.

EQUITY AND UTILITY IN TRIAGE PROTOCOLS AND GUIDELINES FOR ACCESS TO CRITICAL CARE AND VENTILATORS
To illustrate the criteria of equity and utility in protocols in use in North America, we refer to two types of protocols: The New York State Allocation Guidelines (20) (21), from which the Quebec Protocol was derived. These protocols establish the use of common rules for the hospitals and centres that apply them, imposing equal conditions of access to services and thus conditions of impartiality. This procedural equality is expressed in various measures that these protocols and guidelines implement.

Applying the same rules as a condition of impartiality
When the same protocol is applied in all establishments in a region, or even a country, it is an essential condition for equality of treatment for the people living in this territory. If each establishment had its own protocol, there would be no equality of treatment across a territory. The larger the group to which the protocol applies, the more equal it is for all persons requiring treatment. We can speak here of equality before the rule, by analogy to equality before the law (22). In Quebec, for example, if necessary, the Ministry of Health and Social Services can impose the protocol entitled: Priorisation pour l'accès aux soins intensifs (adultes) en contexte extrême de pandémie (23) at the regional level. This establishes a necessary condition for equal treatment for those needing intensive care in the same region. However, it seems that this protocol allows institutions to use different tools, 20 which would introduce a factor of inequality in access to intensive care.
Having a committee that enforces the rules of triage ensures impartiality that clinicians and caregivers may not necessarily have. Many studies show that there are wide variations in physicians' decisions about access to critical care (24)(25)(26)(27)(28), particularly because it is impossible to determine a prognosis with precision and certainty, given that it is based on an assessment of future outcomes that are not all known nor foreseeable. Rather, it is the testing of a treatment that demonstrates its clinical efficacy. Page 11 Another egalitarian procedural measure is the use of an evaluation tool built into the protocol. The New York State Guidelines (20) use the Sequential Organ Failure Assessment (SOFA), and the Ontario protocol (21) uses many tools, such as The Trauma and Injury Severity Score (TRISS), the Pro-Vent Score or the Impact Score.
The very existence of these protocols imposes egalitarian measures for access to care in times of limited resources. This is even clearer when a protocol is imposed on all hospitals by a governmental decree, in a context of scarce or limited resources 21 in terms of intensive care beds and ventilators. In contrast to guidelines, which are more of a guide for health care professionals looking for indications to solve problems created by resource limitations, the decree imposes the application of the protocol. Protocols meet the requirements of equality before the rule that is the first condition of equity. This is not, however, a sufficient condition to promote equity in the allocation of limited resources and access to treatment for those in need. The second condition for equity is that health needs must be adequately met.

Protocols and response to health needs
The most egalitarian criterion of distributive justice in the health domain, compared to criteria based on utility or merit, 22 is treatment based on need. Needs differ from the desires or wishes of the people involved; they must be evaluated by a health professional, a physician in this case. Thus, medical competence is used to make the diagnosis and to judge the appropriateness of using intensive care and ventilators. The inclusion criteria for access to COVID-19 services are distress or severe respiratory failure, and hypotension (12). This assessment, followed by adequate treatment, respects the ethical principles of beneficence, caring and justice in terms of equity. In an equitable hospital setting, all those who meet the criteria for inclusion can access the needed resources, and the medical criteria are egalitarian factors since they are the same for all who qualify. A protocol that allows more room for clinical medical expertise is a better response to individual needs than the intervention of a committee that lacks direct access to patients.

The Ventilator Allocation Guidelines
The triage protocol outlined in the New York State Guidelines (20) proposes a three-step triage process, with a focus on those whose lives are likely to be preserved by the use of intensive care and ventilators. With the goal to save more lives, the first triage step identifies patients who are likely to survive intensive care and ventilator use. Those who would not survive, even with intensive care treatment and ventilator use, regardless of their acute disease, are eliminated; those who do not require intensive care treatment and ventilator use to survive are also eliminated. This first selection avoids overtreatment of people who will die regardless of the means used, and of people who will recover without the use of life-support techniques. It is the health condition assessed by a physician that determines access to intensive care and ventilators, based on an assessment of short-term survival using medical criteria.
The second selection stage assesses mortality risk using the SOFA. This risk assessment tool is based on a clinical examination of the function of six critical organs and systems: lungs, liver, brain, kidneys, coagulation and blood pressure. For each of these organs or systems, four levels of impairment are indicated for a total of 24. If no organ or system is irreparably damaged, the score is 0, compared to all severely damaged organs and systems with a score of 24. The graded ranking corresponds to colours. Red indicates priority treatment, yellow an intermediate level and blue a lower level. Green means that the person does not need a ventilator to survive. There is no reference to patient age in this tool. Subsequent assessments may result in a change in the assessment of priority of access to a ventilation technique. The Triage Officer provides access to treatment using the SOFA score and the medical assessment completed in stage 1. The SOFA score indicates not only the number of organs affected, but also the severity of the damage and the extent of improvement or deterioration, as well as providing an indicator of survival. If many individuals are equally clinically qualified to access intensive care and ventilators, a randomization 23 process is used.
The third stage is an evaluation of the effectiveness of the ventilation technique in terms of whether or not the condition of the person who was able to access intensive care has improved. This evaluation is carried out with the SOFA after 48 and 120 hours of use of ventilation. Compared to the initial assessment, this will be the determining factor in the decision to continue or discontinue ventilator use. Assessments must be made regularly because those whose condition does not improve in real time will have to give up their place to those awaiting treatment.
Assignment or removal of an intensive care bed is done by the triage officer or committee, but clinical assessment is done by the attending physician at all stages of triage; the attending physician assesses the prognosis for survival. Furthermore, the SOFA score is calculated from clinical factors derived from the available medical evidence, not from subjective criteria such as quality of life (20) or from survival statistics by type of disease. In addition, triage decisions take into account all stages of a patient's condition and are based on clinical data and indicators that determine the degree of severity and number of organs and systems affected, as well as the improvement or deterioration of each patient's health with ventilator use (20).

Clinical Triage Protocol for Major Surge in COVID Pandemic
The Ontario protocol does not hide its utilitarian goal: "The overall purpose of a triage system is to minimize mortality and morbidity for a population overall, as opposed to individual mortality and morbidity risk" (21, p.2). It is a solution of last resort, 21 In Quebec, the prioritization protocol is activated when there is an overload of 150 to 200% of intensive care capacity. 22 When it is proposed that caregivers be given priority in the event of a tie in survival assessment, this may be justified in a utilitarian manner (return to work as quickly as possible) or in a way that takes into account their merit for having coped with a difficult work environment. 23 Randomization: random assignment. The protocols studied do not indicate the means by which randomization is performed.
when services can no longer meet the demand. The Clinical Triage Protocol for Major Surge in COVID Pandemic of Ontario, like the Quebec Protocol, is applied according to three levels of selection by exclusion. At the first level, people who are excluded are those whose prognosis of mortality is evaluated at 80% during the following year. If the overload is still too great, the second and third levels are more stringent, eliminating people with a prediction of death > 50% and > 30%, respectively, during the following year.
After the attending physician has assessed the individual health status of those who meet the inclusion criteria (respiratory insufficiency; hypotension), he uses the Triage Protocol to identify patients who would not survive the disease or would only survive it for a short period of time, based on the following exclusion criteria using 13 indicators: 24 1) severe trauma with predicted mortality > 80%; 2) severe burns with any 2 of the following: age > 60 years of age, 40% total body surface area affected, and inhalation injury; 3) cardiac arrest; 4) severe cognitive impairment; 5) advanced irreversible neurodegenerative disease; 6) metastatic cancer; 7) advanced and irreversible immunological impairment; 8) a severe and irreversible neurological event with > 80% risk of death; 9) terminal organ failure; 10) a frailty index ≥ 7 due to progressive impairment; 11) elective palliative surgery; 12) being on mechanical ventilation for 14 days or more with a ProVent Score of 4-5; and 13) clinical judgment indicating a risk of death greater than 80% due to a critical illness or in the near future regardless of the critical illness (21).
With a few exceptions, 25 the Quebec protocol generally uses the same indicators, referring to them as diagnoses and clinical parameters instead of exclusion criteria. In both Ontario and Quebec, the presence of even one of these criteria results in denial of access to intensive care or, in the case of criterion 12, discontinuation of intensive care. In terms of equality in the calculation of survival indicators, the Ontario protocol uses randomization, while the Quebec protocol first uses the life cycle, then priority to caregivers, and finally randomization to choose among candidates with an equal score. The triage levels do not necessarily have to be implemented consecutively and the choice of level is based on the virus outbreak curve.

The age criterion
The age criterion is present in several of the exclusion criteria. The Trauma and Injury Severity Score (TRISS) is designed to assess the survival rate of people with severe and traumatic injuries. It takes into account age, depth of coma (GCS), systolic blood pressure, respiratory rate, and severity of injury. Age is included in the calculation for those 55 years of age and older (29). When the diagnosis is related to severe burns, being aged 60 and older may constitute an exclusion criterion with one of the two following criteria: 40% body surface area, or inhalation burns that result in hypoxemic respiratory failure. Finally, age is related to a frailty score ≥ 7, in the Quebec Protocol, because the scale applies only to those aged 65 years and older in the first level and to those aged 50 years and older in the second level. In these medical scales, age is linked to a severely impaired health condition. It is not used as a simple social criterion.
Unlike the Ontario Protocol, which uses the egalitarian criterion of randomization when the overload is still too high, the Quebec Protocol uses life cycle as a criterion for access to intensive care for those who were not eliminated at the clinical assessment stage, who have a similar prognosis, and who are too numerous for the number of beds and equipment available. The explanation given for the use of this criterion in this protocol refers to "intergenerational equity" presented as individuals' equal right to live their lives. In fact, this practice consists of giving priority to the youngest at the detriment of the oldest; age is thus used as a social criterion of exclusion, and therefore as a discriminatory criterion, in the sense of the Charters. The authors defend themselves by arguing that this is not unjust discrimination because this criterion comes after a first selection stage based on clinical evaluation. However, on one hand, the clinical evaluation is carried out on the basis of exclusion criteria and, on the other hand, whatever the interpretation, age, which is not linked to a health condition, is a discriminatory social criterion even in a pandemic situation. Such discrimination is not justified in the eyes of the law, nor in medical ethics, no more than the other criteria which would be applied subsequently: priority to caregivers (criterion based on utility) and randomization (criterion based on chance). The use of the social criterion of age cannot be ethically justified because it is not equitable: it is not based on need and discriminates against older people who have health conditions similar to younger people, who will be favoured.

COMPARISON OF TRIAGE PROTOCOLS
These triage protocols have similarities. Each has the objective of saving more lives, which in itself is a utilitarian criterion. In all of these protocols, apart from clinicians, intensivists, and nurses at the bedside, a triage committee ultimately decides how to apply the prioritization protocol. According to the University of Pittsburgh Guidelines (30), an independent triage committee aims to promote objectivity, avoiding conflicts of commitment to patients and organization, and reducing moral distress for physicians and caregivers who find it emotionally difficult to stop life-sustaining treatment. In its first version, 26 the Quebec protocol emphasized the importance of saving direct caregivers from the burden of making difficult choices and relieving them of their ethical and legal responsibility in this matter. In the second version, however, it is the attending physician who informs the patient or their next of kin of the decisions of the prioritization committee. In a normal situation, the committee would only be advisory and not decision-making, and the physician would retain full autonomy, rather than simply being the bearer of the 24 The details for each indicator are not mentioned here. 25 Criterion 7 of advanced and irreversible immunological impairment has been removed from the Quebec Protocol. A delay of 21 days instead of 14 days is granted to the person who is on mechanical ventilation and has a ProVent score of 4-5. People waiting for organ donation are not excluded, unlike people who are on oxygen permanently at home. Finally, criterion 8 of the Quebec Protocol corresponds to criterion 12 of the Ontario Protocol, with the difference that a maximum of 21 days instead of 14 is mentioned as the criterion for discontinuing intensive care. 26 The first version published in March 2020 was entitled: Triage pour l'accès aux soins intensifs (adultes et pédiatriques) et l'allocation des ressources telles que les respirateurs en situation extrême de pandémie.
Page 13 committee's decision. Regardless of the protocol used, the physician cannot be legally and deontologically removed from treatment decision-making by a simple administrative document. Any protocol that would deny a physician's decision-making autonomy should be subject to a specific governmental decree.
All the protocols studied use randomization as a last resort. It is an egalitarian and unbiased procedural process, but not equitable, in the sense that it ignores individual needs. Randomization would be ethically unacceptable at an initial triage stage because it would be ignoring the clinical assessment. However, when clinical criteria for access to intensive care have been used in the original selection and there are still too many candidates for the resources available, then randomization, which respects one of the conditions of equity through its impartiality, would be more ethically acceptable.
Prior to randomization, the Quebec protocol takes a utilitarian approach in selecting candidates with similar survival indicators by using the life-cycle criterion that favours the younger candidates. In addition, there is a specific reference to the instrumental value of caregivers who will be given priority in the event of equal scores for candidates for intensive care and ventilators, " to ensure that essential care and services continue to be provided " (23, p.22). Again, this is discrimination on the basis of a social criterion, in this case the type of work. On the one hand, those who might be favoured by this measure could not necessary be in a position to return to work to care for others; on the other hand, why should those who would have access to intensive care be chosen on this basis, since so many health care workers and other types of workers such as care attendants, maintenance workers, and volunteers, play an essential role in managing the pandemic? (20) The Ontario and Quebec protocols are based more on a public health approach, using, among other indicators, data from statistical analyses to assess survival time, taking into account the age of the patients. In contrast, the New York protocol, with the SOFA, is based on individual clinical assessment, without reference to patient age (31). Nevertheless, the New York protocol includes an evaluation of the technical effectiveness of intensive care in real time after 2 and 5 days, whereas the Ontario and Quebec protocols respectively allow a limit of 14 and 21 days for the evaluation of the effectiveness of treatment, at all levels. 27 The New York protocol allows more people to have access to intensive care through its access criteria, but since the evaluation of technical efficiency is done in the short term (2 and 5 days) after admission to the intensive care unit, those who would have benefited in the longer term from intensive treatment are discharged and offered other types of care. 28 It should be noted, however, that a prolonged stay on a ventilator implies serious sequelae (33,34).
Both types of protocols have their limitations. In the opinion of the authors of the Ontario protocol, the third level of triage, which eliminates people with a 30% chance of dying within a year, can hardly be done on the basis of utility. They believe that triage decisions should be based on clinical considerations and not solely on demographic or socioeconomic factors (21). As can be seen, many of the scores proposed by the Ontario and Quebec protocols are based on statistics, that only concern averages. These scores do not reflect the health status of each individual awaiting treatment. Nothing tells us where the individual in question is situated in relation to the average. Only a clinical assessment by a competent physician, based on his knowledge and experience, can evaluate a prognosis which, while still uncertain, has the merit of relating to the particular individual who requires access to care, rather than a curve or an average.
Gabbe and colleagues (35) criticize TRISS for not taking into account the considerable degree of uncertainty in predicting individual survival based on a group prediction. The same can be said for all exclusion criteria that are based on statistics. In fact, in TRISS age is used to reflect individual characteristics (36), with the assumption that older people have more comorbidities, which is a population reality that is not necessarily confirmed at the clinical level for each individual. The same reflection can be made about burnt victims. Excluded are individuals over 60 years of age with burns affecting 40% of the body or inhalation burns; this does not mean that a 62-year-old could not survive with appropriate treatment. Public health compilations are good population indicators, but they have their limitations when applied to triage limited health resources. Moreover, according to the University of Pittsburgh School of Medicine's guidelines (30), the use of exclusion criteria is discriminatory because it eliminates people who would normally have access to intensive care; exclusion criteria are too rigid to accommodate fluctuations in demand during a crisis; and they do not respect a fundamental public health principle of using the means that least restrict individual freedom to accomplish public health goals.
As for the SOFA included in the New York State Guidelines, it is recognized that it is an easy-to-use tool that contains few variables and laboratory tests and whose score is easily compiled by simple addition (20). Instead of proceeding by exclusion, the tool is applied to each individual by adding up the number and severity of impairments, taking into account comorbidities. In this way, it provides a consistent clinical assessment for ventilator allocation. However, it is criticized for not being sufficiently sensitive to differences between candidates with the same status (37), e.g., those with a code red or those with a SOFA score ≤ 7 out of a possible 24. The higher the number, indicating more organs and systems affected, the less eligible is the individual to access intensive care and a ventilator. It is clear that the protocol presented in the New York State Guidelines is more patient-centered and in that sense, adds an equitable perspective in terms of responses to need within a utilitarian structure. However, SOFA bases its assessment on short-term survival, as evidenced by the assessment made after 2 and 5 days of intensive treatment. Some people may be able to improve their condition in the longer term, but they will be discharged and will not have access to treatments that may have allowed them to survive. Those with a high score who do not qualify for 27 These protocols include three levels of triage: see p.7. 28 A study by Wunsch et al. (31) of a cohort of 40,439 individuals who were placed on ventilators for an average of 61.7 hours in 2014-2015 shows that 56.2% were discharged from the hospital. However, this information cannot be generalized to practice in a COVID-19 pandemic, because many factors come into play: severity of illness, sufficient and trained critical care staff, and available techniques. A study by Yang et al (32) conducted at the Jewish General Hospital in Montreal in 2020, including 106 patients with COVID, indicates the use of different techniques, including endotracheal intubation, nasal cannulation ventilation, and prone positioning; the death rate in the intensive care unit (ICU) was 17% which is low compared to the average survival rates in the ICU. treatment, or those for whom treatment will be discontinued because the treatment does not demonstrate efficacy, will be offered other types of treatment and possibly palliative care.
The protocols studied insist on the importance of consulting the patient (or his family in case of incapacity) to make his endof-life wishes known. However, the Canadian protocols studied do not mention the ethical and legal obligations to respect the wishes of a person who refuses the use of a respirator in an Advance Medical Directive (AMD) in Quebec, or who has otherwise expressed his end-of-life wishes, in a mandate or Durable Power of Attorney in Ontario, for example. While there is a legal obligation in Quebec for physicians to consult the AMD Registry, it is possible that in an emergency situation there will not be enough time, and there is great inequality in the consultation of AMD Registry by physicians, depending on their institution and region (38). One of the tasks of the triage committee could be to consult the AMD Registry and the mandate in the patient record. In any case, the prescription of a Medical Intervention Level (MIN) should be made as early as possible in the episode of care and respected by the triage committee. Urgency would not justify a failure to consult the patient or his family in case of incapacity, if time permits, before proceeding with the use of techniques that place a burden on the individual, during treatment and subsequently in rehabilitation.

CONCLUSION
The purpose of this article was to examine triage protocols in terms of equity and utility. Other ethical issues in a pandemic situation, such as the lack of trained critical care nurses, the shortage of staff due to illness, the lack of adequate protective devices, or the comparative effectiveness of treatments and techniques used have not been addressed. What this study highlights is the difficulty of adhering to health ethics principles, including beneficence and equity in the application of triage protocols. It was found that applying protocols creates equality in access to services, but in order for resource allocation to be equitable, it is also necessary to meet health needs, which is opposed to the use of strictly social criteria, such as life cycle or functional utility. However, in times of a pandemic, ethical concessions must be made in order to achieve the goal of public health, that is, saving more lives. It is more equitable to consider not only population needs, but also individual needs. Protocols that balance the two are the most ethical under the circumstances. Analysis of the protocols showed us that the New York State Protocol's use of the SOFA score is more sensitive to individual needs than the Ontario and Quebec Protocols, which have a clear population focus using exclusion criteria based on survival statistics and, as a last resort in the Quebec Protocol, on clearly utilitarian criteria such as life cycle and caregiver priority. From this perspective, SOFA is more equitable in that the triage is based on clinical criteria alone, regardless of any form of social criteria, such as age or employment status.