A Loss of Life Model and Literature Review Contribute to Lessons Learned
As part of a forensic examination to assess the New Orleans flood protection system and its failure following Hurricane Katrina, the U.S. Army Corps of Engineers commissioned a model to calculate potential loss of life from flood events in New Orleans, and a review of extant literature to identify the scope and type of health effects attributed to the storm and its aftermath.
On August 29, 2005, Hurricane Katrina — the third-strongest hurricane on record to make landfall in the United States — devastated New Orleans. The city's protective infrastructure failed as the storm surge breached several canals and levees, flooding over 80 percent of the city and many neighboring parishes. Shortly thereafter the U.S. Army Corps of Engineers (USACE), whose mission includes designing, building, and operating flood control structures, established the Interagency Performance Evaluation Task Force (IPET) to perform a forensic study of the New Orleans and Southeast Louisiana Hurricane Protection System. IPET was tasked with producing comprehensive reports from multidisciplinary analyses of the storm, the flood protection system, the consequences of the disaster, and the estimated risk and reliability of the protection system prior to Katrina, as well as with projecting risk and reliability of the system after the intended repairs and improvements were complete. The USACE was determined to learn the necessary — if difficult — lessons from the disaster and committed to use that knowledge to create a more durable and effective hurricane protection system in the future.
The Corps turned to Abt Associates for two pieces of the IPET appraisal. One Abt Associates team, led by Associate Division Manager Jerry Stedge, was hired to create and test a model to estimate the loss of life related to the flooding; the model and its data would then be used to help assess the risks associated with any future flooding events in the greater New Orleans area. A second Abt Associates team, led by Vice President William Schlenger, was contracted to conduct a comprehensive review of all published media sources, or "gray literature," in order to describe the disaster's real and potential health impacts on the people of the city of New Orleans and the surrounding area.
Stedge's team adapted the LIFESim computer model to address the unique conditions surrounding the flooding of New Orleans.
Stedge's team adapted the LIFESim computer model to address the unique conditions surrounding the flooding of New Orleans. A sophisticated simulation model, LIFESim allows an analyst to estimate loss of life based on catastrophic but relatively brief flooding typically experienced from broken dams. New Orleans was different because there the flooding was comparatively slow but enduring: the water poured into the city as if into a bathtub and was contained for weeks. Further, a dam break is usually a sudden event, in contrast to the significant advanced warning that preceded Hurricane Katrina (which allowed for the evacuation of many potential flood victims). These differences required Stedge and his team to modify how the LIFESim model dealt with water depth and flow velocities, residents' abilities to evacuate the flood areas following warning, and cause-of-death scenarios.
For example, LIFESim originally used estimates based on average fatality rates for each category of flood zone – designated as "safe," "compromised," and "chance." Safe corresponded to a circumstance such as being in an undamaged building above maximum water level; compromised meant a situation such as being in a partially damaged structure; and an example of a chance zone was a location submerged under flowing water. The Abt Associates team modified the criteria to account for the unique aspects of the New Orleans flooding, understanding that most buildings remained intact, but many areas were flooded for extended periods. Abt Associates experts also created an additional, "walk-away," zone to account for areas where people were able to evacuate themselves without assistance. Further, to take into account the disproportionately lethal impact of the flooding on the elderly, Abt Associates experts tabulated impacts on this population using separate criteria for each zone. The LIFESim model, for example, assumes that those under 65 years of age can access attics and roofs, but those over 65 are unable to evacuate vertically above the highest hospitable level.
Many of the LIFESim model parameters — such as rescue efficiencies and fatality rates — were uncertain, so the Abt team used a Monte Carlo modeling technique which allowed for refinement of LIFESim's sensitivity to a range of plausible factors. The team then compared the estimates with verified New Orleans fatalities. The accuracy of the model's results validated its ability to reliably predict loss of life from future hurricanes in New Orleans. Ultimately, Abt Associates experts assessed all 27 New Orleans basins contained within the flood protection system and used criteria from two different sets of conditions: one scenario provided estimates using the population and housing configuration found in New Orleans prior to Katrina in August 2005; the second used population and housing stock projections for New Orleans in June 2006. The second set of results is being used in the analysis of plans for reducing the risk of flooding in New Orleans in the future.
Abt Associates staff on Schlenger's team had to turn to nontraditional sources to provide a literature review that covered Hurricane Katrina's effect on residents' and emergency workers' health. Given that the USACE study began only four months after the disaster, hospital records were spotty or unavailable, and scientific literature was extremely limited, as few studies had yet to be concluded and published in peer-reviewed journals. Schlenger and his team searched newspapers, magazines, websites, and other types of "gray literature" to survey the physical and mental health of those who stayed in New Orleans during the storm, as well as of those who were evacuated, those who served as emergency personnel and first responders, and those who entered the area following the storm as part of the recovery effort. The Abt team created a rich and extensive compilation of anecdotal descriptions of what happened during the storm, and also how the aftermath affected people's health from August 29, 2005 through April 2006, shortly before the report was delivered.
The report on the health effects of the Hurricane Katrina disaster enumerated the different causes of illness and mortality (the latter including deaths from illness as well as from drowning, accidents, homicides, and suicides). Causes included a lack of access to care for those already ill and injured, and not only for those hospitalized and in nursing homes but also for individuals with chronic diseases such as diabetes, which require regular access to medications; infections from wounds and abrasions exposed to contaminated floodwaters, as well as deaths resulting from routine infections that were not treated in a timely fashion; exposures to heavy metals — especially lead, arsenic, and chromium — via contact with soil and sediment and through eating contaminated seafood; respiratory infections, asthma attacks, and other illnesses related to the aerosolized spores from mold; vulnerability to vector borne diseases such as those carried by mosquitoes and from microbial contamination of the floodwaters; and injuries and fatalities associated with electrocutions, smoke, fires, falls, and the mishandling of equipment such as chainsaws used during cleanup operations, to name just a few.
By examining empirical studies of previous disasters such as floods and severe hurricanes, Schlenger's team also quantified the physical and mental health effects of Katrina that could be expected in the near and long term. A significant section of the report discusses the mental health consequences related to Katrina, and describes the psychological trauma and the clinically significant outcomes to be expected, particularly forecasts of the prevalence of different health outcomes such as posttraumatic stress disorder and depression.
Both Abt Associates reports were used by other IPET members to estimate the risks associated with future hurricanes and storms that might impact the New Orleans area. Abt Associates staff are now looking at ways to follow up using data that have been uncovered since the report's publication in May 2006. These experts coordinate with others at Abt Associates who create and implement plans to respond to natural and man-made disasters. This work ranges from finding ways to evacuate hospitals and nursing homes safely ahead of a severe storm, to anticipating emergency medical care necessitated by a large-scale terrorist attack. The people of Abt Associates continue to look for lessons to learn from prior disasters so that similar ones can be avoided in the future.
