Memorial Medical Center patient evacuation triage

Memorial Medical Center patient evacuation triage was the improvised system used to sort and move patients as the hospital prepared to evacuate during Hurricane Katrina. It converted scarce space, transport, and time into rankings that determined who would leave first and who would wait. The system was formed under extreme conditions, without a settled protocol or meaningful patient participation; its treatment of patients with Do Not Resuscitate orders made the human consequences of that ranking especially clear.

From emergency bands to numbered groups

At about seven in the morning, Memorial’s incident command team planned to consolidate patients and supplies in staging areas near exits on the first and second floors. A nurse was sent to retrieve the hospital’s color-coded emergency triage armbands: green for patients who could walk, yellow for those needing assistance, red for those dependent on others for care, and black for patients with DNR orders. The bands were meant to make patient needs legible as people were moved through the building.

The hospital could not locate enough of the official armbands. That shortage prompted a replacement system for the more than one hundred patients remaining from Memorial and LifeCare Hospitals. Staff assigned them numbers according to their anticipated evacuation order: “1” for patients in relatively good health who could walk or sit up, “2” for those who were sicker and required more help, and “3” for those to be evacuated last. The final group included patients judged very ill and, under an earlier decision, patients with DNR orders. Nurses reviewed charts in dim light, conserving flashlight batteries, then wrote the category on paper and taped it to patients’ clothing or gowns.

The groups were also physically separated. Roughly three dozen patients categorized as 1 were brought to the emergency-room ramp, where LifeCare nursing director Gina Isbell and Memorial nurse manager Karen Wynn supervised their care while they waited for boats. Perhaps seventy patients categorized as 2 were placed along a corridor leading toward a hole in the machine-room wall and a route through the parking garage to the helipad. Around eighteen patients categorized as 3 were gathered in a corner of the second-floor lobby, near a bank ATM and a planter. The labels were not merely an administrative shorthand: they determined where people waited and how soon they might reach a way out.

The DNR decision and its meaning

The decision to rank DNR patients last had been established by Memorial doctors before the numbered system was introduced. Four patients in the intensive care unit had DNR orders, including Jannie Burgess and Helen Breckenridge. Richard Deichmann, the medical chairman who later described his understanding of the decision, believed that the law required DNR patients to have a certified terminal or irreversible condition. He thought they had the least to lose if calamity struck and should therefore go last.

But a DNR order had a narrower meaning: if a patient’s breathing or heartbeat stopped, the patient should not be revived. It was not the same as a living will, which under Louisiana law could address the withholding or withdrawal of life-sustaining procedures in specified circumstances. DNR status did not itself mean that a patient had agreed to forgo evacuation. The difference became consequential when a ranking that concerned rescue was attached to a medical instruction about resuscitation.

At Memorial, decisions about categorization were made by medical professionals alone. There was no system for explaining the rankings to the patients and families most affected, and in some cases staff deliberately withheld the reason for delay. On a fifth-floor corridor, transporters stopped moving a bed-bound patient after discovering her DNR order and repeatedly maneuvered other patients around her. Nurses were told not to tell her daughter that the order was why her mother had not been taken downstairs. The daughter had agreed to a DNR order because she did not want her ninety-three-year-old mother revived after cardiac arrest; she had not understood it to mean that her mother would be left behind during evacuation.

When the daughter learned what was happening, she asked for the order to be canceled. A doctor was found to remove it, and the patient was moved downstairs toward the helipad and rescue. Her immediate ordeal ended in survival. On the seventh floor of LifeCare Hospitals, Angela McManus likewise sought to rescind the DNR order for her frail seventy-year-old mother, Wilda, after hearing that DNR patients would be evacuated later. Staff told her no doctor was available to do so. Wilda remained upstairs. When doctors at the staging area instructed LifeCare staff to stop sending patients down because there was no more room, the ranking system’s practical limit became explicit: placement in a category could not guarantee that the next stage of evacuation would be available.

A system under pressure

The numbered triage took shape amid a larger failure of evacuation capacity. New Orleans residents without cars faced an evacuation order without adequate transport, while ambulance dispatchers were overwhelmed with requests from hospitals, nursing homes, and homes. Many ambulances were caught in traffic; some patients were taken to the Superdome rather than out of the city. Inside Memorial, staff were contending with darkness, failed communications, and deteriorating conditions as they tried to move patients toward limited exit points. Anna Pou joined the work of organizing the mass movement, helping search for cots and stretchers and receiving patients brought down from other floors.

The historical purpose of triage is to allocate limited resources when the number of people needing help exceeds what rescuers can provide. But the book emphasizes that assigning a low priority is not a neutral or reliably predictive act. Assessments of who is most likely to survive are uncertain; even experienced rescuers can disagree, patients’ conditions can change, and additional resources may become available. Some approaches therefore argue that patients thought unlikely to survive should still receive care or evacuation after those needing immediate intervention, but before those able to wait. At Memorial, staff had little if any training in triage systems and no particular protocol guiding the improvised categories. Pou later described the sorting as heart-wrenching and understood the shift from evacuating the sickest first to placing them last as an expression of the doctors’ belief that they could not save everyone.

That change also shaped the meaning of the numbers. A category intended to speed evacuation could appear to declare a patient’s fate, especially when it was assigned without consultation or explained to the person waiting. Neuroradiologist Bill Armington regarded the method as expeditious but distasteful. The system’s compression of a patient’s circumstances into a single digit made movement easier to organize, but concealed uncertainty and could obscure the difference between needing more assistance and being beyond help.

Consequences and lasting memory

The triage decisions remained part of the emotional burden carried by staff after the evacuation. Gina Isbell later described poor sleep, nightmares in which she saw the faces of her patients, and guilt about the people who had died and the workers she had selected to work through the hurricane. Even after the special grand jury’s decision in 2007, she remained angry that help had arrived too slowly and felt that the storm had changed her permanently. She also held to the belief that she and her colleagues had done the best they could, and remembered a patient’s grandmother thanking her during the final hours on the helipad.

Karen Wynn’s recollections likewise joined the work of evacuation to its aftermath. Preparing for a possible grand-jury appearance meant revisiting events with her lawyer; afterward, she sometimes needed weeks to feel able to move forward again. She recognized what she believed were lasting signs of post-traumatic stress in herself and colleagues and said she no longer wanted the responsibility she had carried for years as a nurse and manager. These consequences do not settle whether the triage decisions were justified. They show instead that the categories remained more than a temporary logistics measure: they became part of how the people who made and witnessed them remembered the crisis.

The episode reveals the limits of treating disaster triage as a purely technical exercise. Memorial’s staff had to make urgent choices amid scarce transport and failing infrastructure, but the choices also depended on judgments about illness, rescue, and whose account of a patient’s wishes would be heard. The DNR cases exposed the cost of allowing a medical status to stand in for consent to delay or deny evacuation. In the hospital’s improvised ranking, a number could help organize a route to safety—and could also conceal how much remained unknown about the person waiting beneath it.