Article Text

Download PDFPDF

Developing the duty to treat: HIV, SARS, and the next epidemic
  1. J Dwyer1,
  2. D F-C Tsai2
  1. 1
    SUNY Upstate Medical University, Center for Bioethics and Humanities, Syracuse, NY, USA
  2. 2
    Department of Social Medicine and Department of Family Medicine, National Taiwan University College of Medicine, Medical Humanity Building, Taipei, Taiwan
  1. Dr J Dwyer, SUNY Upstate Medical University, Center for Bioethics and Humanities, 725 Irving Avenue, Suite 406, Syracuse, NY 13210, USA; dwyerja{at}


SARS, like HIV, placed healthcare workers at risk and raised issues about the duty to treat. But philosophical accounts of the duty to treat that were developed in the context of HIV did not adequately address some of the ethical issues raised by SARS. Since the next epidemic may be more like SARS than HIV, it is important to illuminate these issues. In this paper, we sketch a general account of the duty to treat that arose in response to HIV. Our purpose is not to defend or criticise this account, but to show that it needs to be developed in order to address three important issues. The first issue concerns how risks should be distributed among healthcare professionals. The second issue concerns the conflicts that arise between professional duties and family duties. The third issue concerns the forms of support that societies owe healthcare workers during epidemics. Our descriptions of these issues are drawn from our experience of the SARS epidemic in Taiwan.

  • duty to treat
  • HIV
  • SARS
  • epidemic

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.

SARS, like HIV, placed healthcare workers at risk and raised issues about the duty to treat. But philosophical accounts of the duty to treat that were developed in the context of HIV did not adequately address some of the ethical issues raised by SARS. Since the next epidemic may be more like SARS than HIV, it is important to illuminate these issues. In our paper, we shall sketch a general account of the duty to treat that arose in response to HIV. Our purpose is not to defend or criticise this account, but to show that it needs to be developed in order to address issues about the distribution of risks, family duties, and social support. Our description of these issues is based on our experience of the SARS epidemic in Taiwan.1


The spread of HIV in the 1980s occasioned a serious and fruitful discussion about the duty to treat. Because some healthcare workers avoided treating HIV+ people, the healthcare professions were forced to discuss questions about social roles, ethical duties, and the moral meaning of their professions. This discussion was especially intense in America, where ideas about a doctor’s right to select patients were more widely accepted and even incorporated into codes.

Ezekiel Emanuel, John Arras, and Norman Daniels contributed to the discussion by offering accounts of the duty to treat.2-4 These thinkers approached and grounded the duty in slightly different ways, but their views had much in common. Because they saw that a simple appeal to historical records and professional codes was inadequate, they looked to the idea of social roles and practices. Although people have some duties simply because they are human beings or because they make promises or enter into contracts, they also have duties because of the social roles they assume. Parents, teachers, and firefighters have duties associated with their roles. These duties may entail actions that involve inconvenience and risk. Firefighters, for example, have a duty to put out fires and save lives even at some risk to themselves. When they become firefighters, they accept a certain amount of risk that is associated with this role. Individuals are free to reject this social role and choose a safer occupation, but they are not free to reject all risk within the occupation. In other words, they are not always free to separate and select particular duties that are bundled in a given social role.

What’s true of firefighters, teachers, and parents is also true of physicians, dentists, and nurses. Consider the profession of medicine. At its core, medicine is about treating and caring for ill people. This is the ideal inherent in the practice. By choosing to become doctors, people accept some duty to treat. Although individuals may tailor their practices in certain ways, they may not refuse to see a patient simply because the patient has an infectious disease. The issue here is not a peripheral matter left up to individual choice, but a central concern that helps to define the role of doctoring.

Although Edmund Pellegrino takes a somewhat different approach to the issue, he too focuses on duties that are embedded in the role and practice of medicine. He describes essential characteristics of the practice of medicine and then appeals to obligations that are internal to this practice. In his view, these obligations require putting aside self-interest and accepting risk.5

In the account that we have sketched, accepting a certain kind and amount of risk is seen as an essential, inherent, or defining element of the practice of medicine. But shouldn’t we be a little sceptical about claims that a particular duty or virtue is a defining element of a complex practice? Yes, but this scepticism need not undermine the ethical point. We can grant that the ethical duties and ideals that help to characterise the profession of medicine are always somewhat imprecise and open-ended. We can accept that the ethical core is open to debate and still assert the ethical point.

The view that the normative core is revisable can be incorporated into the account that we sketched. The claim about the duty to treat need not be based merely on an analysis of the concept of medicine or the ideals of the practice. It can also be based on the presentation of a social choice about the kind of persons, roles, and societies we want to work to create. A society like ours could choose to let the duty to treat erode away from what we count as doctoring, but that choice is not a good one. It has too many disadvantages for potential patients, social institutions, and doctors themselves. Or so one could argue.3

Although some social roles require people to accept a reasonable level of risk, these roles do not require an absolute disregard for personal safety. Firefighters are not required to enter, and are ordered out of, burning buildings that are about to collapse. But what risks are reasonable and what risks are beyond what duty requires? That probably depends on the benefits involved, the level of risks, the importance of conflicting responsibilities, and the ethical expectations in certain roles and fields.

Because the level of risk matters, many commentators tried to determine the risk of occupational transmission of HIV. After considering the number of patients who are HIV+, the frequency of exposures like needle sticks, and the rate of sero-conversion, many commentators concluded that the risk to most healthcare professionals was within the reasonable range. Indeed, the estimates in the 1980s were often higher than the actual risks proved to be. Although Emanuel, Arras, and Daniels did not think that the line between reasonable and excessive risks could be drawn with precision, they viewed HIV as a case in which the risk is not too much to ask.


Although the account that we sketched did not specify an exact cut off between duty and supererogation, it did help to focus attention on the issue of reasonable risk. At the beginning of the SARS epidemic, no one knew the exact risk. Early evidence indicated that about 30% of cases were healthcare workers and about 10% of all cases were fatal. Since SARS posed a substantial occupational risk, people debated whether the duty to treat encompassed this risk.

But even if we have a clear account of what counts as a reasonable risk, we still need to address the issue of how the risks should be distributed. At least in the USA, the risk associated with caring for HIV+ patients was very unevenly distributed. Some cities had much higher rates of infection than others. Within most cities, the risk fell disproportionately onto staff at public hospitals and clinics. Within hospitals and clinics, the risk was higher among some specialties than others. And within specialties, the risk tended to shift down the hierarchy, to residents and nurses. Part of the problem was personal and professional: some healthcare workers did not affirm a duty to treat, and they shifted a burden onto others. But part of the problem was structural. Because the USA lacked a system of universal coverage, patients who had no insurance turned to public hospitals for care.

Both Arras and Daniels noted that the distribution of risk was unfair. The picture of fairness that is operating here is that all professionals should be willing to do their share, and that the shares of risk should be more even. This made sense for HIV, but SARS was different. It would be a public health disaster to spread SARS patients evenly throughout practices, clinics, and hospitals. With diseases like SARS, it makes sense to cluster patients in designated units and hospitals. But this means that the associated risks will also be clustered. Although this kind of clustering serves a legitimate public health purpose, it also raises an issue of fairness.

Even when everyone accepts a duty to treat, the problem is to arrange responsibilities in a way that is both fair and efficacious. Consider some of the arrangements that need to be made to deal with a disease like SARS.6

  1. Public health officials need to direct people with symptoms to particular clinics and hospitals. For example, in large cities with many hospitals, officials may designate certain hospitals to handle most of the patients.

  2. Hospital administrators need to arrange emergency screening in a safe and efficacious way. For example, an administrator might assign one area of the emergency department and one subset of the staff to screen patients with fevers.

  3. Hospital administrators need to designate wards for people with confirmed cases, suspected cases, and possible cases.

  4. Hospital administrators may need to reassign healthcare professionals from one area to another.

All these decisions and arrangements raise issues about what is fair and reasonable.

Appealing to an abstract duty grounded in professional commitment does not adequately address the problem of distribution. It is not our purpose in this paper to try to solve the problem, but we want to point out one direction to explore. Perhaps the problem calls for a more democratic approach. It is easy to caricature and ridicule a call for more workplace democracy, especially in a field like medicine. But we believe that an ethical response to the problem of distribution requires that the stakeholders have meaningful input in the development and implementation of safe and fair arrangements. Since most of the arrangements could be made well in advance, there is often adequate time. What we lack is not time, but habits of management and democratic forms of deliberation that ensure proper consultation, consideration, and respect.


During the SARS epidemic, even more than during the HIV epidemic, healthcare workers experienced a conflict between their professional duties and their family duties. Many workers felt obligated to treat infectious patients, but they are also felt a responsibility to respect their parents, protect their spouses, and care for their children. Conflicts like these are a common feature of moral life because people have multiple roles.

When faced with a conflict of duties, it makes sense to try to determine which duties have a greater claim and what factors should be given greater weight. Ezekiel Emanuel, for example, recognises the conflict and resolves it in this way:

As a general rule, when the risks of contracting AIDS are not excessive by accepted social standards, the professional obligation must be fulfilled, regardless of the physician’s marital status– just as firefighters and soldiers with families must still risk their lives to fulfill their professional obligations when those risks, although real and high, are not deemed excessive.2

Given the general low risk of exposure to HIV, Emanuel believes that the professional duties take precedence.

We do not disagree with Emanuel’s conclusion, but we see the need to supplement the approach of weighing conflicting duties with an approach that seeks to accommodate competing claims. In his essay on “The Moral Philosopher and the Moral Life,” William James talks about the need to fashion inclusive arrangements in which opposing demands are met to the highest degree possible. He explains his view in a highly metaphorical way:

That act must be the best act, accordingly, which makes for the best whole, in the sense of awakening the least sum of dissatisfactions. In the casuistic scale, therefore, those ideals must be written highest which prevail at the least cost, or by whose realization the least possible number of other ideals are destroyed. Since victory and defeat there must be, the victory to be philosophically prayed for is that of the more inclusive side,– of the side which even in the hour of triumph will to some degree do justice to the ideals in which the vanquished party’s interests lay. The course of history is nothing but the story of men’s struggles from generation to generation to find the more and more inclusive order. Invent some manner of realizing your own ideals which will also satisfy the alien demands,– that and that only is the path of peace! Following this path, society has shaken itself into one sort of relative equilibrium after another by a series of social discoveries quite analogous to those of science.7

James’ approach calls for a greater reluctance to sacrifice competing demands and a greater commitment to trying to discover or invent more inclusive arrangements.

We do not know of any simple way to reconcile professional demands and family demands in times of epidemics, but we do believe that better social arrangements could ameliorate the situation. Here are three concerns that arose in Taiwan during the SARS epidemic.

  1. Protection. Healthcare workers were concerned to protect their families against infection. To do so, they took a variety of steps. Some wore masks at home and avoided hugging and kissing their children. Some slept on a couch in the living room. But those healthcare workers who wanted to protect their families by not going home had to scramble to find accommodations. Surely hospitals and societies could find creative ways to help. Medical centres could make rooms and dorms available. Societies could provide healthcare workers with hotel vouchers.

  2. Childcare. Healthcare workers, like most people, were deeply concerned about care for their young children. When a resident in the emergency department had to work longer hours, and was later quarantined at the hospital, she had to find extra childcare for her year-old daughter. Many people experienced similar problems. They had to work longer hours, deal with baby sitters who quit, and assuage the fears of childcare centres. Societies need to recognise this conflict as an ethical problem and to fashion better arrangements.

  3. Education. Healthcare workers, like most parents, were concerned to provide their children with a good education. During the SARS epidemic, many of them worried that they would die and be unable to fulfil this responsibility. If parents die, there is no way to replace their unique and personal influence on their children’s education, although other people may step forward to help. But societies can address financial concerns with mechanisms like scholarships and survivor’s benefits.

Societies, ministries of health, medical centres, and various associations could fashion arrangements that would reduce some of the tension between professional demands and family demands.


In the previous section, we considered some social arrangements that could help healthcare workers deal with competing demands. In this section, we want to reflect more generally on the kinds of social supports that would be appropriate. What does society owe healthcare workers during epidemics?

First of all, societies and hospitals should take reasonable steps to ensure the safety of healthcare workers. The account of the duty to treat that we sketched makes this point in a clear and direct way. Emanuel, for example, states that it “is incumbent on society and the medical profession to reduce the risk” to healthcare professionals who treat AIDS patients.2 In his reflections on the lessons of SARS, he writes: “Affirming healthcare workers’ ethical duty to care for the sick imposes a correlative duty on healthcare administrators and senior physicians to quickly develop and deploy procedures to maximise the safety of frontline physicians and nurses.”8

We certainly agree with this point, but we would also emphasise the need to provide appropriate forms of social support. Healthcare workers, like other workers, need health insurance, disability insurance, child-care, survivor’s benefits, and so on. These benefits are important because of what they provide but also because of what they express: social appreciation of the risks that healthcare workers take. Society needs to find ways to express respect, fairness, solidarity, and related notions. Although this is very important to do, it may be difficult. Consider a few examples from Taiwan.

During the SARS epidemic, the Taiwanese government promised and later delivered extra pay for healthcare workers who treated SARS patients. Doctors who treated SARS patients received extra pay of about US$300 per day. Nurses received extra pay of about US$150 per day. This form of support is problematic for two reasons. Because it looks more like a form of incentive pay than a recognition of an important social role, it may work to undermine a sense of duty. And because the differential pay for doctors and nurses did not correspond with the risk of exposure, the arrangement seems unfair.

Appropriate benefits are important forms of social support, but we also need to consider less tangible and more symbolic forms of support. How should the community acknowledge the commitment and sacrifice of people who serve a vital role during an epidemic? Consider one bad example. During the peak of the SARS epidemic in Taiwan, a nurse worked her usual shift, caring for patients. When her shift ended, she walked over to the bus stop near her hospital. Dressed in her uniform, she waited patiently for the bus. The bus arrived but passed her by because the driver and the people on the bus wanted to avoid any risk of infection. Shunning healthcare workers when the risk is relatively small shows a lack of appreciation and solidarity.

We would even expand the discussion of support to include the idea of meaning. Because social roles are bound up with duties, benefits, and meaning, we want to consider the social support of meaning. When we interviewed healthcare workers, we gave them a chance to express a variety of concerns. Although they noted problems about the lack of proper equipment, the distribution of risks, the difficulty of reconciling professional duties and family duties, and the lack of community support, what troubled them the most were the shortcomings that hindered patient care.

A chief resident in the emergency department complained about how difficult it was to transfer patients. When there was no bed available at her hospital for a patient who probably had SARS, she called two other hospitals. They said they had no room. She suspected otherwise. Another resident was troubled by what he heard about a patient he had cared for. This patient died after he had been moved to another unit. The resident heard that because the patient had SARS, he was not intubated in a timely manner. These doctors were trying to realise certain ideals of care, but the realisation of these ideals depends on social cooperation, organisation, and conditions. There is a lesson here for us. We shouldn’t think of social support merely as external goods that are owed healthcare workers for the role they assume during epidemics. We should also think of social support as a means to help healthcare workers realise goods, ideals, and meanings that are internal to the practice of healthcare.

Something that two physicians said reminded us of the importance of internal goods. Both of them were very involved in caring for SARS patients. One was a specialist in infectious diseases. He worked long hours caring for patients, consulting on cases, designing protocols, and supervising treatment plans. The other was a chief resident in the emergency department. She worked long hours screening patients, providing care, and placing patients. During the middle of the epidemic, both of them were quarantined and separated from their families for 14 days. At the end of an interview, conducted in English, one of us asked them if there was anything else they wanted to say. They hesitated, looked at each other in a sheepish way, and whispered something to each other in Chinese. Then one of them spoke up: “Yes, we wanted to tell you that we miss that SARS time. We were real doctors then.”


We are grateful to the Chiang Ching-kuo Foundation for a grant that supported this work, and to Department of Social Medicine of the National Taiwan University College of Medicine for other support. We would also like to thank the following people for their comments and encouragement: RA, JMA, DB, KC, YC, M-SH, WH, H-CN, RLP, WR, WR, and an anonymous reviewer. We would especially like to thank all the doctors and nurses who talked to us about their experiences, concerns, and views.



  • Competing interests: None declared.