Memorial Medical Center's Hurricane Katrina Crisis exposed how a foreseeable infrastructure failure could become an evacuation emergency and, ultimately, a lasting dispute over patient care, responsibility, and the meaning of emergency medicine. The danger was not simply that a major storm would overwhelm the hospital: staff had examined how flooding could disable essential systems, while the crisis unfolded amid disrupted communications, delayed rescue, and reports about the outside world that were not always reliable. The later investigation did not settle whether the disputed medication of patients constituted murder, and the psychological and professional consequences extended well beyond the evacuation.

Foreseen Vulnerability and Rising Floodwaters

The hospital’s vulnerability had been examined after Hurricane Ivan narrowly missed New Orleans in 2004. Eric Yancovich studied storm-surge models with the U.S. Army Corps of Engineers and calculated that a severe storm could bring twelve to fifteen feet of water around Memorial, which stood three feet below sea level. His hand-drawn elevation sketch showed that less than four feet of street flooding could carry water over a loading dock and into the hospital. Although the backup generators were on the second floor, essential parts of the emergency power distribution system were below ground or only a short distance above it.

Yancovich warned that relatively little flooding could disable most of the Medical Center and predicted that patients might have to be moved to the newer Clara Wing, whose electrical circuitry he believed was better protected. Memorial had added floodgates and raised a vent for its underground diesel tanks, but his recommendations to raise emergency power switches and pumps supplying medical air and suction remained unfulfilled. A partial bid for electrical work exceeded a quarter of a million dollars; his memo attributed the delay to a lack of capital. As floodwaters approached, the weakness was therefore not an unforeseen technical detail but a risk identified in advance and left only partly addressed.

The crisis also took shape through an increasingly unreliable information environment. Hospital staff heard a neighboring parish president announce on WWL that martial law had been declared, but it had not been declared in the disaster zone. The station also broadcast alarming claims about looting and violence. At the same time, callers described the continuing flood: one man, Freddy, reported that he and four others, including a baby, were trapped in an attic in the Gentilly neighborhood, surrounded by about nine feet of water as they tried to signal a helicopter. Such reports conveyed genuine peril, but other dramatic broadcasts blurred the line between confirmed events and rumor.

Isolation and the Evacuation Perimeter

Inside Memorial, the command team believed that generator power might last only hours. The hospital was not sealed off from people seeking help: a stabbing victim waded to its emergency-room ramp, and families arrived by boat with children, looking for food and assistance. Helicopters carrying people away were visible to those outside, attracting others in desperate circumstances. Security concerns led staff to secure exits, while hospital personnel considered directing families to the Superdome, more than two miles away. Reports on WWL, however, suggested that conditions there were also dire.

The broadcasts continued to shape how staff understood the city. LifeCare nurses heard reports of hostage situations, prison breaks, shootings, and armed looting. Some of the claims were uncertain even as they were transmitted; a reported shark in a hotel area, for example, might have been debris. The broadcast atmosphere could make the surroundings seem more comprehensible by giving them a narrative of lawlessness, while also increasing fear. In the hospital, actual needs remained immediate: patients required care and transport, and staff had to make decisions while communications and outside assistance were uncertain.

Medication, Investigation, and Unresolved Responsibility

The most consequential dispute concerned medications administered to patients on September 1 as boats and helicopters emptied Memorial. Ethicist Arthur Caplan argued that the reported circumstances did not fit the usual safeguards for palliative sedation. If the purpose had been solely to ease pain and distress, he expected care in the pace of medication and documentation. Medication orders had been recorded by hand earlier in the disaster, suggesting that recordkeeping, though difficult, was possible. He was also troubled by the apparent rapid administration of substantial amounts of drugs known to be lethal, the absence of prior use in some patients, and the lack of consultation with family members who were present. He did not accept that killing was the only way to relieve suffering.

Frank Minyard’s own judgment remained divided. He said that, under such conditions, he would have tried to do what he thought best for each person and would have placed sick people before those who were not sick. At the same time, he said he would not have killed patients because they could not be removed from the seventh floor. He did not claim that Anna Pou had done so. He considered some of the deaths potentially classifiable as homicides, but could not make the same determination for every patient, whose illnesses and susceptibility to medication differed. Emmett Everett stood apart in his assessment: Minyard regarded Everett’s death as unequivocally caused by drugs. The case remained unlike his usual work, in which causes were clear; at Memorial, evidence and interpretation continued to pull against one another.

A special grand jury declined to indict in 2007. The decision lifted some of the pressure on LifeCare staff, but it did not persuade every juror that no crime had occurred. One juror later recalled the patients who had been alive in the morning and dead by afternoon, particularly Everett, whom she remembered eating breakfast and asking when the staff would begin. She believed a crime had taken place, even though the jury returned no indictment. The difference between that conviction and the legal outcome captures the unresolved evidentiary problem: the jury did not hear testimony from a witness who said she had seen Pou inject a patient.

Trauma and the Public Account

The crisis persisted in the lives of those who had worked through it. Gina Isbell described poor sleep, nightmares featuring patients’ faces, and guilt about the staff she had selected to work during the storm. She had left Memorial by small boat and then learned that her home and community in St. Bernard Parish were lost. Even years later, she wore a hurricane-shaped pendant and remained angry that help had arrived too slowly. A memory from the helipad offered a counterweight: Roy Culotta’s eighty-seven-year-old grandmother told Isbell that if she lived until the next day, it would be because of her. Isbell continued to believe that she and her colleagues had done the best they could and would serve again in a storm if asked.

Karen Wynn’s aftermath included preparing for possible grand-jury testimony with her lawyer. Revisiting the events could leave her struggling for weeks afterward, and she recognized what she believed were symptoms of post-traumatic stress in herself and colleagues. She no longer wanted the responsibility of managing nurses or choosing staff for a storm. John Thiele, who later survived a prolonged critical illness, described frightening distortions during sedation. A nurse who had cared for him believed he seemed remorseful about the patients he had helped inject; Thiele later rejected that interpretation. He said his own condition differed because its problems were potentially reversible, a distinction he considered central to how he understood the patients at Memorial.

Pou later became a national lecturer on ethical questions in disaster medicine. In those talks, she presented an account that differed from the documented record: she said officials had told LifeCare to remove patients by noon, described a shortage of clean water despite bottled water being left after the evacuation, and placed Coast Guard helicopters at the hospital only late Thursday afternoon. She did not discuss the patient injections, although she used her arrest to argue for legal immunity for medical professionals in future emergencies. The omission mattered because it framed her public case around difficult triage and disaster conditions without addressing the action at the center of the investigation.

The Crisis Revisited

The book’s later return to the hospital, by then known as Ochsner Baptist, places the crisis beside evidence of repair and continued care. In 2012, as Hurricane Isaac approached, the chapel and corridors recalled the deaths and evacuation seven years earlier; Cheri Landry and Lori Budo were again caring for patients in the surgery building. That return does not resolve the old dispute. Instead, it makes continuity visible: the same medical setting can hold both renewed care and a history that remains painful to those who worked there.

The crisis also raises a practical question broader than the disputed decisions at Memorial: what prevents an emergency from forcing clinicians to choose among inadequate options? A later analysis in the book argues that saving lives depends not only on refining triage rules, but on reducing the need to compromise care in the first place and restoring resources as quickly as possible. At Memorial, the plan to medicate patients proceeded at a moment when helicopters were arriving in greater numbers. The contrast underlines the importance of situational awareness and repeated reassessment: emergency conditions can change, and decisions made as if resources were fixed may fail to account for help that is arriving.