Inadequate emergency preparedness at Memorial Medical Center

Inadequate emergency preparedness at Memorial Medical Center became decisive when Hurricane Katrina exposed a vulnerability hospital leaders had identified but not remedied: floodwater could disable the electrical systems on which patient care depended. The hospital had extensive emergency plans, regular committee meetings, and a formal preparedness process. Yet the plans did not fit together, the hazard assessments overstated readiness, and essential infrastructure remained exposed. As conditions deteriorated, staff were left to improvise around a foreseeable threat.

Known Flood and Power Vulnerabilities

The danger was not wholly unforeseen. After flooding disabled power at hospitals in the Texas Medical Center in 2001, New Orleans health director Kevin Stephens asked the city’s hospitals whether they could withstand fifteen feet of water, what it would cost to raise generators, and whether they wanted the city to explore federal funding. The response was unenthusiastic: such work could cost millions of dollars, and the initiative went no further. In the days before Katrina, federal emergency officials were told that all but two of the city’s roughly eighteen hospitals had generators, electrical switches, or both at ground level. Memorial Medical Center was among the exposed hospitals.

After a near miss in 2004, plant operations director Eric Yancovich attended a meeting about storm-surge models and learned that a Category Four or Five storm could bring twelve to fifteen feet of water around Memorial Medical Center, which stood three feet below sea level. He then sketched the elevations of hospital entrances and outdoor equipment on lined paper, using Magnolia Street as his reference point. His calculations showed that less than four feet of street flooding could send water over the loading dock and into the hospital. Although the generators themselves were on the second floor, parts of the emergency power distribution system sat below ground or only inches to a few feet above it.

Yancovich warned his supervisor that relatively little water could disable most of the medical center. He expected that the main hospital would lose power and that patients would have to be moved to the newer Clara Wing, where he thought the electrical circuitry was better protected. His department installed floodgates and raised the vent for underground diesel tanks, but the more extensive fixes remained undone. Yancovich had recommended raising emergency power transfer switches and the pumps supplying medical air and vacuum suction, which respiratory patients relied on. A partial bid for electrical work exceeded a quarter of a million dollars. Citing the lack of capital, he kept the proposal on file for future consideration.

Plans on Paper, Gaps in Practice

Memorial Medical Center’s emergency planning process was extensive in form. Each year since 2001, Susan Mulderick convened a preparedness committee to assess about forty-seven possible hazards using a three-page template. The committee produced scores for probability, risk, and preparedness, but its latest matrix contained numerous multiplication errors. Members rated readiness for power outages, generator failure, and flooding as good, the highest available category. They later said they had focused more on combinations of emergencies that had occurred before than on the worst events that might occur.

The plans also reflected an imbalance in what the hospital prepared for. Its most detailed emergency scenario addressed bioterrorism and ran 101 pages; the hurricane plan was eleven pages. Memorial Medical Center maintained twenty separate plans totaling 273 pages, but they did not provide a usable response to the crisis that unfolded. The hurricane plan did not anticipate flooding, the flooding plan did not account for the need to evacuate, and the evacuation plan did not anticipate losing power or communications. There was no guidance for a complete power failure, no plan for evacuating through flooded streets, and no arrangement for helicopter transport. The hurricane plan assumed the generators would operate for at least seventy-two hours, even though they had never been tested for that duration.

The committee’s approach also made self-assessment stand in for a practical test of capability. Unlike some hospitals, Memorial Medical Center had not hired an outside consultant; the committee used a freely available online form while under pressure to control costs. Years later, Bob Wise, who had helped introduce national emergency-management standards at JCAHO, criticized the collection of plans as disconnected and too numerous for anyone to master. In his view, Memorial Medical Center had treated accreditation paperwork as a substitute for all-hazards preparation: documenting responses mattered more than ensuring the organization could carry them out.

The Crisis Exposes the Gaps

When word arrived that floodwater could reach fifteen feet, the emergency meeting at Memorial Medical Center took place in a hot command center where a metal oxygen tank had been used to break windows. The hospital had ranked hurricanes, floods, and power outages among its highest-priority hazards, but its plans did not join those risks into a single scenario. Staff faced a combination they had not been prepared to manage: rising water, threatened power, and the possible need to evacuate nearly two hundred patients. Mulderick proposed moving the sickest patients first, including intensive-care patients, newborns, high-risk pregnant patients, dialysis patients, and bone-marrow-transplant patients.

The generators and the systems connecting them to the hospital became an immediate test of the preparations. The three teal-colored generators, each rated at 750 kilowatts, had already been running for two days. The maintenance staff’s routine work consisted mainly of oil changes and brief monthly tests; none of the electricians and engineers on site was a generator mechanic. When one engine overheated, workers added water to its radiator and restarted it. Later, suspecting low oil pressure, they brought diesel to prime the motors. As outages spread and lights dimmed, staff confronted an electrical system made complex by decades of additions to the original 1926 hospital.

After one generator failed, workers tried to connect some of its load to another. Wearing rubber boots, they waded through knee-deep basement water to reach the electrical building’s mezzanine. About a third of the automatic transfer switches appeared to be at a low level and potentially submerged, along with associated distribution panels—precisely the kind of vulnerability Yancovich had warned about. Workers used a bypass handle to shift the load, but a gauge then indicated that the working engine was drawing too much current. They tried to reduce the load by shutting off circuits to unused equipment and fixtures. Another generator failed; lacking spare parts, the team could not repair it by scavenging components. At about two in the morning on August 31, nearly forty-eight hours after Katrina made landfall near New Orleans, the last backup generator surged and died.

Evacuation Beyond the Hospital

The hospital’s preparedness failures formed part of a wider evacuation problem. New Orleans Mayor Ray Nagin ordered an immediate evacuation, but many residents had no car or other way to leave. Officials knew transportation plans lacked enough drivers, while people who could evacuate independently were already stuck in traffic. The Superdome was designated a shelter of last resort, and people with medical conditions were directed there without an explanation of what care would be available. Dispatchers at Acadian were overwhelmed by calls for transportation from hospitals, nursing homes, and homes; many available ambulances were trapped on the interstate, and some began taking patients to the Superdome rather than out of the city.

The city had also rehearsed a catastrophic hurricane through the Hurricane Pam exercise, whose scenario assumed more than two thousand hospital patients would be in New Orleans. But no one had worked out how to move so many patients to safety through a flood, and federal health officials had not joined the latest planning sessions. At Memorial Medical Center, boats eventually carried some people to dry ground, but evacuation remained a shifting, constrained operation. Staff tried to manage patients while also confronting uncertainty about where evacuees would go and what conditions awaited them. One morning, an outpatient reached the hospital through floodwater for a scheduled chemotherapy appointment, an arrival Karen Wynn recalled as a sign that people outside the building did not understand the emergency within it.

Consequences and Significance

The disaster’s costs accumulated across interconnected vulnerabilities: an evacuation order that came late, inadequate transportation, disjointed rescue efforts, poor communications, and hospitals and nursing homes that had not evacuated before the storm or invested in backup power and water systems robust enough for a prolonged emergency. Memorial Medical Center’s own record sharpened the significance of those broader failures. It had identified the elevation risk, considered repairs, and rated its preparedness highly, yet it had neither completed critical upgrades nor built a coherent plan for the combined threat of flood, power loss, and evacuation.

The account does not reduce the crisis to a single defective machine or a single decision. Floodwater threatened low-lying switches; prolonged generator operation strained equipment designed for shorter use; and the hospital’s plans did not provide staff with a workable route through the resulting failures. The contrast between the paper assessment and the practical emergency is central: a high preparedness score could not raise a switch above flood level, supply missing expertise or parts, or arrange transport that had never been secured. At Memorial Medical Center, the first real test of the plans became the event they had failed to prepare the hospital to meet.