New York City serves as a second major setting for the book’s examination of hospital disaster response: during Hurricane Sandy, the danger of a storm-driven power failure again placed critically ill patients and the decisions made on their behalf under urgent pressure. The city’s experience at Bellevue Hospital Center follows the New Orleans crisis by only two months, making it a test of what had—and had not—been learned from Hurricane Katrina.

A new storm, familiar vulnerabilities

On the night of October 29, the narrative moves from the hurricane response in New Orleans to a meeting at the North Shore-LIJ Health System command center on Long Island. Officials there were coordinating support for their own hospitals and responding to calls for help from elsewhere in the region. New York University’s Langone Medical Center, in New York City, was seeking ambulances during the storm to evacuate four critically ill babies from its neonatal intensive care unit.

The city’s exposure was shaped by its evacuation policy as well as by the storm itself. Mayor Michael Bloomberg had exempted hospitals in low-lying Zone A from the pre-storm evacuation order, as Ray Nagin had exempted hospitals in New Orleans before Katrina. City and state health officials knew that some hospitals’ backup generators could fail if high tides and storm surge reached them. The parallel is not exact, but it makes the central risk plain: keeping a hospital open can leave patients and staff inside a facility whose essential systems depend on vulnerable infrastructure.

Bellevue’s power threat

At Bellevue Hospital Center in Manhattan, a storm surge driven up from the East River was forecast to flood the area around the hospital. Hospital leaders warned Laura Evans, the medical director of critical care, that flooding could disable both the pumps supplying oxygen and vacuum suction and the pumps moving fuel from underground tanks to backup generators. If city electricity failed, the generators’ small attached fuel reserves would last only about two hours.

The stakes were concentrated in Evans’s fifty-six-bed adult intensive care unit. Nearly every patient depended on electrically powered equipment; many were receiving life support or intravenous medicines to regulate heart rate and blood pressure, and some relied on intra-aortic balloon pumps. The threat was therefore not simply a building losing power: a failure could interrupt the machinery sustaining patients whose immediate survival depended on it.

Before the city utility supply failed, Evans and her colleagues prepared to distribute the six available power outlets. They made a patient list, moved patients between beds and outlets, and communicated their choices to other staff; the process took about two hours. Evans wanted decisions that could be explained, applied consistently, and defended as fairly as possible. She knew the story of Anna Pou and Katrina and understood that the choices might later be scrutinized by the public as well as by the hospital.

The contingency becomes real

As the team completed its list, the lights went out. Backup power came on after about seven seconds, but the command center soon warned that all power except at the six outlets could be lost within sixty to ninety minutes. Floodwater was filling the basement of the twenty-five-story hospital and entering elevator pits with enough force to dislodge elevator doors.

Staff prepared for the possibility that powered equipment would stop. Two health professionals were stationed beside each ventilator-dependent patient to provide oxygen manually with Ambu-bags. Bright-orange extension cords from the backup generator system ran through the corridors, while nurses counted drops on intravenous pumps so they could administer vasopressors manually if the pumps failed. These preparations made the crisis visible in small, practical measures: a limited number of outlets, cords crossing patient areas, and staff ready to replace automated support by hand.

Lessons and limits

The contrast with Katrina is significant but incomplete. The regional command center described in the account was organized and offered proactive assistance to hospitals beyond its own system, unlike the belated support the book says Tenet and LifeCare had provided during Katrina. Yet some staff still had not planned for a complete failure of backup power—the failure of the backup plan itself. New York City’s emergency underscored that improved coordination could not eliminate the deeper uncertainty of deciding how to distribute scarce life-sustaining resources.

At Bellevue, the difficulty extended beyond choosing who would receive an outlet. Evans and her colleagues hesitated to tell patients and families who would not receive power, and delayed communicating the triage decisions. In retrospect, Evans believed families had a right to know. The staff also did not consider asking whether a selected patient or family might volunteer to give up an outlet for someone else. The New York episode thus shifts the book’s inquiry from infrastructure alone to the procedures and relationships through which emergency decisions are made: a plan may be deliberate and transparent to staff, yet remain incomplete if those most affected are not informed or invited to participate.