NFPA 99: Standard on Health Care Facilities

NFPA 99: Standard on Health Care Facilities is a national standard governing health-care facilities, including requirements for their electrical systems. Its significance in the book lies in the gap between a code-compliant system and one able to keep a hospital functioning through an extended disaster. Memorial Medical Center’s loss of backup power makes that distinction concrete: the generators were only one part of a vulnerable system, and their brief routine tests did not demonstrate that the whole system could withstand flooding and prolonged use.

Scope and Adoption

The National Fire Protection Association publishes the national electrical code, including specific requirements for health-care facilities. NFPA 99: Standard on Health Care Facilities had been adopted by most state health departments and licensing agencies, giving it a broad role in the regulatory framework for hospitals. It is distinct from NFPA 110: Standard for Emergency and Standby Power Systems, which addresses emergency power systems specifically.

The distinction matters because emergency power depends on more than the generator itself. At Memorial, the generators stood above the water, but parts of the distribution system—including transfer switches and panels—were at low elevations. The book’s account therefore places NFPA 99 within a wider network of facility standards and operational dependencies rather than presenting it as a guarantee that every component will remain usable in a catastrophe.

Testing and the Limits of Compliance

At the time of the crisis, hospital generators had to be tested monthly, but only for short periods. Memorial’s maintenance staff typically ran tests once a month in the middle of the night. The generators were treated much like heart-lung bypass machines: equipment expected to support vital functions for minutes or hours, even though Memorial’s disaster plan called for them to sustain the hospital for at least three days. By the time the system failed, the generators had already been running for two days.

This mismatch exposed a limitation in what routine testing could establish. A short monthly test checked that equipment could operate briefly; it did not recreate the demands of continuous operation during a prolonged outage, or prove that the connected electrical infrastructure would remain available in flood conditions. The generators’ breakdown at Memorial thus raised a broader standards question: whether a system that meets ordinary testing requirements is prepared for the conditions it may face in a major disaster.

Memorial Medical Center and the Failure Chain

Before Katrina, Memorial’s plant operations director Eric Yancovich had studied storm-surge models and mapped hospital entrances and critical equipment against the elevation of Magnolia Street. His calculations indicated that less than four feet of street flooding could send water over the loading dock. Although the backup generators were on the second floor, important parts of the emergency power distribution system were at or near ground level. Yancovich warned that relatively little water could disable most of the medical center and recommended raising vulnerable switches and pumps; the proposed electrical work was not approved because of limited capital.

During the crisis, one generator stopped, and workers restarted it after addressing overheating. Later, faced with low oil pressure and other failures, they tried to redistribute electrical loads, working through water in the basement and reaching a bypass handle near electrical panels. Some transfer switches and distribution panels appeared to be submerged. The effort to connect loads to a functioning generator produced signs of excessive current, while another generator failed and could not be repaired. After roughly two hours of troubleshooting, the last backup generator surged and died at about two in the morning on August 31, 2005.

The precise cause of the shutdown was disputed: an expert for the plaintiffs attributed it to floodwater shorting transfer switches and panels, while an expert for Tenet attributed it to mechanical problems in the generators. The episode nevertheless showed that the hospital’s resilience depended on a chain of components, some of which had been recognized as flood-vulnerable before the storm. A generator’s location above floodwater could not by itself protect the system that distributed its power.

Revisions and Continuing Questions

After Katrina, The Joint Commission warned that compliance with NFPA codes and standards was only a starting point and that systems meeting those standards might still prove insufficient in major catastrophes. In 2007, it required emergency generators to undergo a continuous four-hour test once every thirty-six months, in addition to the existing monthly tests of thirty minutes. These changes extended testing beyond the short monthly runs described in the account, but they did not erase the broader concern about prolonged outages and connected system vulnerabilities.

The book also notes that hospital and nursing-home generator failures during later prolonged utility outages—including Hurricanes Gustav, Isaac, and Sandy—suggested that many systems remained inadequate. NFPA 99’s role is therefore presented neither as a single cause of Memorial’s failure nor as a complete remedy. It belongs to a standards framework whose practical effectiveness depends on testing, facility design, investment, and attention to local hazards—especially the parts of an emergency power system that can fail even when the generators themselves remain above water.