By Philip M. Rosoff, MD

In the late summer of 2005, Dr. Anna Pou faced a horrendous and unanticipated moral and medical dilemma.

An attending otolaryngologist at Memorial Medical Center in New Orleans in the immediate aftermath of Hurricane Katrina, Pou had no electricity to power the apparatus of modern medical care, with the emergency generators having failed. Caring for large numbers of patients whose very lives depended upon those machines, and with no rescue in sight, she and the few colleagues who remained behind had to decide who should live and who should die.

Which of the acutely and chronically ill patients could be carried down the stairs to the helicopters and boats that might not arrive in time? Who should be permitted to suffer in unspeakable agony as the hours went by without any sign of relief?

Ultimately, she decided that she would ease their suffering using the only tools she had available -- morphine and midazolam. And for that, she was excoriated in the press and accused of murder by the state.

What happened after Hurricane Katrina was a national tragedy in which thousands died and many more lost their homes and livelihoods. Compounding the disaster was the poor preparation of the medical community for such an overwhelming catastrophe, with physicians like Dr. Pou being forced to make decisions for which they were prepared poorly, if at all.

While it might be tempting to argue that Katrina was a once-in-a-lifetime event, and that it’s improbable that doctors would have to cope with such a situation again, that seems like a meager rationalization for failing to prepare for another disaster.

Most recently, many doctors, nurses, hospital administrators, public health experts, and medical ethicists have been discussing and formulating plans for how to confront a possible influenza pandemic that could paralyze the medical system for months at a time, inundating health care facilities and providers with incredibly sick patients.

While the outcome of the 2009–10 flu season was milder than many expected, there is no reason to expect that we will not face such a situation in the not-too-distant future, or that the discussions that took place were in vain. Indeed, many of the topics that were the subject of argument and deliberation taught us to scrutinize what medicine can and can’t do, what we owe patients, and consequently what we should do in a crisis.

In other words, when we can’t try to save everyone, how do we decide whom we should attempt to save, and what, if anything, do we owe those who “lose” the lifesaving lottery? The answers to these fundamental questions stretch our moral selves to the limit. They force us to address issues that we rarely take time to think about in this country: concerns about rationing medical care and whether there are some lives that are more important than others.

Over the past couple of years I have given a number of lectures about this topic, using pandemic influenza planning to illustrate the problems. To demonstrate the profound difficulties in making decisions such as these, I have used the following scenario and table to challenge the audience in their thinking about how they would decide who lives and who dies if forced to make a choice.

Before reading the table below, imagine that you are in charge of triage at Duke Emergency Department (ED) during a flu pandemic. Many people, young and old, are presenting with incipient respiratory failure.

The ED is inundated with patients. Under normal times, when intensive care unit (ICU) beds and ventilators are almost always available, there is little discriminatory thinking required: if someone can physiologically benefit from mechanical ventilation, even in the short-term, and if there is not a valid do-not-attempt-resuscitation order in place, we intubate and resuscitate.

But what do you do when you have two patients in the ED at the same time whose vital signs are virtually identical, but you only have one ventilator and ICU bed available?

Who gets the chance to live? Who is relegated to die?

Who Will Live and Who Will Die?

If you are like most people, these questions and their possible answers make you extremely uncomfortable. Not only does the way we address them reveal possible inner prejudices and potential cracks in our carefully crafted moral personae, but also demonstrates that we don’t have a clue about how we might act if we were in Dr. Pou’s shoes.

This thought experiment demonstrates the singular importance of open and inclusive discussion before such a crisis presents itself, so that the answers that are provided by the consensus plan do not compel doctors and nurses to face the choices in the table alone and unprepared.

Without proposing specific answers to these quandaries, we are not left without recourse. We can -- and indeed, should -- create plans that consider these horrendous scenarios and offer supportable and justifiable reasoning for choosing some people over others that are both fair and, as much as possible, evidence-based.

In point of fact, in the United States and in many other countries, a number of plans to deal with pandemic influenza have been crafted. Although most of these plans do not have specific advice on these questions, some do, for instance those of the VA system and the nascent plan created by the North Carolina Medical Society. Common features of these plans have included mechanisms to medically justify clinical decision making, using evidence-based medicine whenever possible.

Both the plans and the rationale for them need to be publicly vetted and acceptable, especially those parts that may have the most controversial, perhaps even noxious, statements. Given the uproar that accompanied the recent health care reform legislation, with talk of “death panels” and the like, it is vital that the process be transparent.

Furthermore, when adopted, the plans must affect everyone: it is a good practice of public justice that those who make the rules should also be bound by them. Hence, the relative of the hospital CEO does not have a greater claim on intensive care resources than someone who is possibly sicker and more likely to benefit. It may also be reasonable to advance the idea that some members of society are special, not necessarily because of who they are, but for what they do.

For instance, one could make a plausible argument that first responders and others who might justifiably profess to play vital roles in a health care catastrophe, such as police and firefighters, National Guard troops, ambulance drivers, doctors and nurses, and power plant workers should have first claim on such resources as influenza vaccine (as indeed was the case last year).

Finally, the primary mission of health care is the relief of suffering. And there is no question that a disaster such as a pandemic or another Katrina or a major terrorist event would produce human suffering on a massive scale. In the event of not being able to save everyone, we should prepare to care for those who are relegated to go without lifesaving treatment. This means training a wide variety of personnel in basic palliative, end-of-life care and stockpiling the drugs and other supplies that would be required to comfort the dying and relieve their suffering.

There is no guarantee that such plans will work as intended. But it is guaranteed that without a plan, human misery and anguish will be widespread, and moral principles that we rightly hold dear would be violated.

Careful discussions before the fact that include as many voices as possible will maximize the chances of how we, as a society, could successfully meet a prolonged emergency with our moral dignity both intact and strengthened.

Philip Rosoff, director of clinical ethics at Duke University Hospital, is a pediatric oncologist with a master’s degree in philosophy.

Duke’s clinical ethics service provides education to physicians, nurses, and house staff; generates policy and policy changes regarding ethical issues; and offers mediation to patients, families, and physicians when questions arise regarding such issues as end-of-life care, patient ability to participate in decisions, and futility.