In the late summer of 2005, the waters loosed in the wake of Hurricane Katrina laid siege to New Orleans. At Memorial Medical Center, the power and then the backup generators failed, creating a silence one doctor described as the "sickest sound" of his life. Doctors and nurses, at times in darkness, struggled to take care of patients without life-saving machines, air conditioning, or functioning toilets. After several days of desperation, some allegedly euthanized critically ill patients, even as large-scale evacuations of the hospital began. The journalist and doctor Sheri Fink published a meticulous investigation of these deaths in theNew York Times Magazine and on the Web site of ProPublica, in 2009. Her work won a Pulitzer Prize.
And now comes the book. In "Five Days at Memorial," the contours of the story remain the same, yet Fink imbues them with far more narrative richness, making the doctors seem both more sympathetic and more culpable. Fink also expands on the ethical conundrums, which have festered over time and seem to gain fresh urgency whenever the angry winds of a hurricane crash against our shores. Which patients should receive priority in a disaster? Should doctors abandon the critically ill? What should they do if they believe those patients will suffer? Should they receive legal immunity for decisions made under duress? To a disturbing degree, the best answers seem to run counter to what key players at Memorial did and what at least one of them has advocated for since.
As the waters surged toward Memorial, hospital leaders began by acting with familiar priorities: they chose to evacuate the sickest patients and those who relied on ventilators or other devices first. But then the medical chairman decided that patients with do-not-resuscitate orders should get lowest priority, later saying he thought they had the "least to lose." As the staff scrambled to move patients down from the upper floors, other factors played a role. One doctor worried that a patient who weighed more than three hundred pounds might slow the evacuation line; he was moved to another area to wait, rather than joining the queue. Some patients on the seventh floor seemed to get even lower priority, in part because their treatment was overseen by LifeCare, a health-care company that leased space within Memorial. As the days passed and doctors soldiered on without electricity, running water, sleep, or outside help, they flipped their moral scale upside down. Now the sickest patients—whom they designated threes on a one-to-three scale—would wait, and the healthiest would go first. This reflected "a sense among the doctors that they would not be able to save everyone," writes Fink, citing the head-and-neck surgeon Anna Pou.
Setting the right triage priorities is one thing. But then something darker began to happen. Four days into the ordeal, Dr. Ewing Cook, who'd previously survived two heart attacks, struggled to reach the eighth-floor intensive-care unit, where he examined a patient with advanced cancer. Convinced that she was close to death and that he couldn't possibly make it up the stairs again, he asked the nurse to raise her dosage of morphine. "There's no question I hastened her demise," he later said. Cook allegedly spoke with Pou about injecting patients with morphine and the sedative Versed, and she wrote prescriptions for substantial amounts of morphine, which, despite the chaos, were filled by the hospital pharmacy. (Whether her intent was to relieve pain or end life remains in dispute.) Meanwhile, the pulmonologist John Thiele apparently injected several category-three patients with morphine and Versed. Knowing that he would soon evacuate, he felt he could not "justify hanging a morphine drip and praying it didn't run out after everyone left and before the patient died," Fink reports.
There are moments when doctors face genuinely tough calls about whether to let someone live or die: a soldier on a battlefield, say, who begs a doctor not to leave him behind alive. But this isn't the way Fink describes the scene at Memorial. Some of the patients whose bodies contained high levels of morphine or Versed were apparently not on the verge of death or even in terrible pain, according to details of the investigation. A three-hundred-eighty-pound paraplegic named Emmett Everett had eaten tuna fish, crackers, and relish for breakfast. "I knew he was sick," one staff member later said, "but, um, you know, he could talk and everything." In fact, he told one of his nurses, "Cindy, don't let them leave me behind." Then a doctor allegedly ended his life—without his consent. Even after the earthquake in Haiti, physicians from a field hospital said they did not amputate survivors' limbs without consent, regardless of how urgently they thought patients needed the procedure. Ultimately, over twenty of the bodies found at Memorial contained either morphine and Versed or both. Multiple forensic experts determined that eight or nine of these deaths were homicides. Pou was arrested, but a sympathetic grand jury, which for political reasons was not privy to all of the forensic evidence, failed to indict her. No other doctors or nurses were held to be criminally responsible.
The more gritty detail Fink offers, the harder it is not to feel sympathy for the doctors and nurses. Yet much of what they did—especially the euthanasia but also the triage priorities—sets an awful example for future disasters. For starters, a D.N.R. order simply means that "a patient whose heartbeat or breathing had stopped should not be revived," Fink writes. It reflects that patient's wishes for the future; it does not necessarily indicate how sick he is now, or how close to death. Using D.N.R. orders in triage also risks penalizing individuals who plan for scenarios that may not occur. At Memorial, the early reliance on these orders meant that some patients were never carried down from the seventh floor to the staging area for evacuation. This made it harder to include them later on, even as more helicopters arrived.
A triage plan that puts the most vulnerable last is also troubling. Doctors may reason that the very sick are too fragile or too heavy to transport or don't have long to live. But they're the ones who are most likely to die if left in the hospitals as conditions get worse. A system that prioritizes the healthy—or the unhealthy—should also be looked at with humility. Death is unpredictable, and experts aren't necessarily good at differentiating among critical patients. Indeed, in a small exercise designed to plan for pandemic flu, researchers asked doctors to assess the likely survival of I.C.U. patients with the H1N1 virus who were on ventilators. Their predictions, as Fink notes in her epilogue, were largely wrong.
Some doctors have been criticized for leaving Memorial, even when food, water, and pain medication were still plentiful. But with gunshots in the distance and rumors of violence and martial law, it seems reasonable that some of them chose to evacuate. "They are not the Secret Service," the bioethicist Arthur Caplan told me. It also makes sense that they would try to lessen their patients' pain beforehand, even if the drugs they used risked hastening death. If their intent was to end suffering forever, though, this crosses a line, Caplan and others argue. After all, who can know whether some unforeseen rescue might have materialized at the eleventh hour, however unlikely it seemed?
Pou and her team have argued that we should not second-guess doctors and should instead grant them broad legal immunity for their work during disasters. But this is neither necessary nor wise. Doctors are already judged according to the resources available and the circumstances, as George Annas of Boston University told me. The standard of care is what "a reasonable physician would do in that disaster, not in a fully equipped O.R. in New York." And the last message we should want crisis workers to receive is that anything goes.
Terrified, weary doctors should not abandon ethical standards; they should cling to the norms of medicine, beyond when it feels reasonable to do so. This means talking to patients and respecting their autonomy, even under awful conditions. It also means making every effort to evacuate the sickest and most dependent patients first. The original—deeply reasonable—triage plan at Memorial prioritized babies, I.C.U. patients, those who required dialysis, and those who had received bone-marrow transplants (and therefore had compromised immune systems). If hospital leaders had been forced to choose some of these patients over others, it would have been reasonable to send babies over the elderly, since infants arguably have the most life ahead of them. (Alternatively, some ethicists propose simply drawing straws.)
But the grim hypotheticals risk obscuring another issue: medical workers under siege can easily lose perspective. They can start to make decisions based on their own dark fears rather than the changing facts on the ground. "For heavens sake," one of the forensic experts cited by Fink said, "Memorial wasn't on a goddamn battlefield with enemy shells coming in. This was New Orleans, and there were helicopters and boats. And really, were they saying they couldn't get patients off the seventh floor?" In any disaster, medical teams must reëvaluate patients' conditions and the available resources frequently, as Fink argues, "maintaining the ability to 'see' in the midst of a crisis." This, along with better backup generators, is what may ultimately save the most lives.
Amanda Schaffer is a science and medical columnist for Slate and a freelance science writer.
