By Wendy Marie Thomas, American Meteorological Society, Policy Program
As Earth’s climate system changes, weather patterns, which are subordinate to climate, are subject to alterations. To protect our critical infrastructures, in particular, and society, in general, it would be helpful to know what the weather of the future will look like. Will we have more frequent storms, more intense storms, or will former stormy areas experience more tranquil weather?
Determining these changes with accuracy is complex, primarily because one input (e.g., temperature rise) can yield singular or multiple outputs (e.g., deeper storm convection, wind flow changes, and precipitation changes) over a variety of temporal and spatial scales. Researchers in hydro-meteorology are investigating whether the anthropogenic (or human) component is leading to a stormy world. Some argue that these influences to climate have already contributed to higher losses, and this point is used as an indicator that severe weather patterns are increasing. 1, 2, 3 Others, on the other hand, say that recent losses are the result of population and economic growth. 4, 5, In essence, this argument holds that people and our buildings are moving into the path of storms, thus giving the appearance that more storms are occurring.
As these investigations play out in science journals and discussions, there are some factors that decision-makers in the United States (from the community to federal levels) need to consider in recognizing our nation’s vulnerability to weather extremes, most especially to our critical infrastructures. These following considerations are realities of today and the foreseeable future, and are reason enough to spur mitigation and adaptation for critical infrastructures, most especially hospitals – the focus of this article.
The United States, primarily a large continental mass, exhibits a multitude of regional weather patterns (e.g., Atlantic hurricanes, Midwest tornadoes, Pacific Northwest precipitation, and Southwest monsoon rains). In terms of extreme weather, the nation has the highest number of tornadoes of any country in the world. 6 Also a reported 55% of the total population lives within 50 miles of a coastline. 7 Combined with population growth, significant concentrations of our society are exposed to flood risk. More concerning is that floods are currently the leading weather-caused fatality in the United States (based on 30-year average and 2009 and 2010 fatality data). 8
With respect to health care critical infrastructure (e.g., hospitals and clinics), a Federal Emergency Management Agency (FEMA) report cited U.S. Census data that showed an increase in new hospital construction by 65% between the years 2000-2006 9. This growth likely mirrored a real estate boom and expansion of the population into new spaces closer to weather risks. Moreover, building codes, though varying at the state and municipal levels (even for critical infrastructure), tend to be more stringent for earthquakes than severe weather because the codes are created based on the probability of destruction. According to the Insurance Institute for Business and Home Safety (IBHS), a direct hit from an earthquake is more widespread and potentially more destructive (having a 1 in 500 or 1 in 1000 chance of occurring per year). In contrast, the spatial generality of storms (e.g., projection for hurricane landfall in the Gulf, versus hurricane landfall in Mobile, Alabama) reduces the chances of a direct hit. More to that point, tornadoes, which can occur in a narrow area and indiscriminately strike buildings and people have a 1 in 5,000 chance of happening.10 The low odds of a direct hit to any one building or a community, even in tornado alley, do not cultivate the urgency or need to bring buildings up to a higher code, for to do so would bear exorbitant upfront economic costs.
Thus the combination of current weather realities, population expansion, and infrastructure vulnerability in the United States are cause enough to warrant effective mitigation (i.e., damage prevention) disaster risk reduction strategies. The potential that climate change can aggravate existing conditions only gives more energy to the urgent need to protect society from a range of potentials.
To take a closer look at how these realities bear on health care infrastructure, it is useful to take a geospatial view of the United States’ healthcare infrastructure. In the visualization we can see layers of society and pieces of this segment. We can identify the location of vulnerable populations, the nation’s 6,300 individual hospitals, numerous clinics and private doctor offices that together form the “housing units” for our healthcare system.
In this geospatial view, we can also see the environmental risks, like severe weather and seismic activity that can damage or destroy these essential facilities. In tranquil times they are a source of health and medical attention, a hub of employment and services (e.g., contributing 17.6% to the GDP of the U.S. economy). 11
During weather (and other) disasters they are critical arms to the response. If standing, they are a sanctuary for the ill and injured. If damaged or destroyed, they amplify the disaster response as attention is diverted from patient care to patient and building care. Moreover, depending on the size and severity of the impact, neighboring hospitals become affected through massive patient surges to emergency rooms. Finally, after a disaster, hospitals are among the critical infrastructures that can make or break decisions and the momentum to rebuild. 12 Recent events reveal that our health care infrastructure is vulnerable to severe weather.
These stories highlight that when a hospital building fails, so too does the continuum of health care delivery. People (e.g., patients, staff, and families) suffer, and this is the primary reason why a closer look at risks and mitigation opportunities is needed for hospitals.
Some Recent Events:
EF-5 Tornado, St. John’s Hospital in Joplin, Mo.
Event: On May 22, 2011, Joplin, Mo., was struck by an EF-5 (winds greater than 200 mph) causing 150 fatalities and 1,000 injuries. The Joplin tornado is now ranked among the deadliest in the United States since record keeping of tornadic activity began in 1950.
Hospital damage: St. John’s hospital, a nine-story facility, was severely damaged. The hospital windows were blown out, triggering a massive evacuation of 183 patients and 200 staff. In addition, gurneys were tossed up to five blocks away, and x-ray films were found as far as two counties (~70 miles away). St. Joplin’s patients were transported to a makeshift hospital in an open field across the street, and to a hospital in Springfield, Mo. (~70 miles away). 14
EF-3 Tornado, Sumter Regional Hospital in Americus, Ga.
Event: At 9:20 p.m. on March 1, 2007, an EF-3 (winds 136-167 mph) struck Taylor and Sumter counties, where Americus, Ga. (within Sumter) was the hardest hit. Americus is located 135 miles due south of Atlanta. For the total affected areas, two confirmed deaths and numerous injuries were reported. 15
Hospital damage: Sumter Regional Hospital was the leading facility for the area. It is located 70 miles away from the nearest full-service hospital, and serves a rural population of 16,000 with a 143 bed facility. The EF-3 tornado severely damaged the building by blowing out the front wall and windows. Glass and debris projectiles became embedded into the walls. Rain also infiltrated the exposed building, and flooded the basement where medical records were kept. These records were permanently lost. No fatalities were reported at the hospital. However there were several injuries that resulted from the storm and the rescue mission, which occurred late at night.
In addition to damage to Sumter Regional Hospital, the surrounding dental and medical office buildings and clinics within a 3-mile radius, which formed a network of health care for this town, were also destroyed along with medical records. The loss of health care facilities forced residents to travel as far as 70 miles to the nearest medical facility. This posed a problem for those with chronic (e.g., cancer, dialysis, diabetes, etc.) and acute (e.g., injuries, cardiac arrest, stroke, etc.) pathologies.
After the storm and during the recovery period, staff reportedly suffered signs of Post-Traumatic Stress Disorder (PTSD), and thus were not as functional as before. 16
Hurricane Ike, Houston, Texas
Event: On September 13, 2008, Hurricane Ike made landfall in Texas as a Category 2 storm. An estimated 112 people were killed in the United States. Two days before landfall, an evacuation order was issued for Galveston and Houston, and many residents fled to the Dallas/Ft. Worth area.
Health Care Issues: The University of Texas Medical Branch in Galveston was severely flooded, resulting in an estimated $710 million to $1.4 billion in damages. Meanwhile further north, in Dallas, chronically ill patients who fled from the storm in the south experienced breaks in needed healthcare that put many in a more severe health risk.
Maryanne Kridner, RN & Diabetes Nurse Educator from Baylor All Saints in Dallas/Fort Worth was on the scene, and described to a group of meteorologists, and emergency managers that disasters put the chronically ill at greater risk of danger for several reasons.
She says that panic causes people to flee without their life-saving medicine, lack of refrigeration leads to medicine spoilage, trauma can induce psychological shock (therefore, requiring higher levels of care), and family separation lead to a lack of support that could otherwise uplift a person’s hope and confidence that make medical care more effective. Chronically ill patients (such as those with cancer and diabetes) receive the medical treatment and care at “Medical Special Needs” (MSN) shelters. However, there is a tendency during mass evacuations for people (who are fleeing homes) to go into general shelters, which tend to lack the proper medication (which is critical for diabetics). This was the case during Hurricane Ike.
California Wildfires, Providence Holy Cross Medical Center, Burbank, Calif.
Event: In 2008, two California wildfires affected southern California: The Sesnon Fire (a natural fire, October 2008, lasted 5 days) and the Sayre Fire (an arson, November 2008, lasted 6 days). The Sesnon Fire destroyed 17 residences, 63 other buildings, and resulted in $12.6 million in damages. The main threat from this fire to Providence was smoke, not fire. The Sayre Fire was more intense. It jumped a highway, which complicated the response as it prevented needed staff from reaching the facility. The fire caused $13 million in damages, including 480 mobile homes (including three homes of hospital staff).
Hospital Damage: Providence Holy Cross Medical Center — the only local area trauma center in Burbank — received more than 200 patients from neighboring hospitals and canceled all elective surgeries. Providence was able to stay open and operational during both fires, in large part, due to the use of HEPA filters. These filters were purchased for pandemic flu preparedness to purify the air and to support the central ventilation system for maintaining zero pressures (which is critical for quarantine rooms). De-pressurization of quarantine rooms would otherwise require an evacuation because of the public health risks. In this case, the utilization of equipment intended for one purpose actually helped the hospital remain open and functional during a fire/weather event. 17
Iowa Floods, Mercy Medical Center, Cedar Rapids, Iowa
Event: In June 2008, historic floods affected much of the Midwest: Indiana, Illinois, Iowa, and Wisconsin. The event resulted from a mixture of above-average winter precipitation (2007-08) and a strong stationary front in the spring of 2008. Over-saturation caused several waterways to overflow, including three major rivers: Des Moines, Cedar, and Wisconsin. The massive flooding washed out key infrastructures, including 125 miles of primary highway. The total costs are an estimated $10 billion in damages. 18
Hospital Damage: In Cedar Rapids, (one of the hardest hit areas), Mercy Medical Center was one of its two main hospitals (located only a few blocks from each other) that had to be evacuated. Pressure from the floods caused a sewage backup. The hospital staff had to go on diversion status, meaning they had to work to evacuate all 205 patients and personnel (representing ~85% of full capacity). The evacuation was further complicated by the fact that not all staff could reach the hospital. Thus a limited staff that had been on duty for more than 24 hours worked to transport chronically and acutely ill patients. 19
Event: Despite 10 years of warning from the hydro-meteorological community that a storm similar to or more powerful than Hurricane Katrina would strike New Orleans, La., the buildings and most of the local population were ill-prepared for the powerful storm that reached land on Aug. 25, 2005. 20 The category 3 storm (by landfall) caused a levee breach, which in turn led to massive flooding that affected five hospitals. A delayed response meant that patients and staff functioned on rationed food and water supplies, with make-shift sanitary operations for waste removal for 3- 5 days.
Hospital Damage: The impacts from Hurricane Katrina on the hospital and public health systems were many. Looking only at the massive hospital evacuation from New Orleans, we can see the impacts from an “imported surge”— a high-volume influx that usually emanates from a catastrophe at a distant location. Emory University Hospital in Atlanta, Ga., was one of the primary receiving units, which took in 300 patients who were air-evacuated two at a time from New Orleans.
The “imported surge” was problematic for both the patients and the receiving unit. Patients lack a continuum of care, and this can be dangerous for those with chronic diseases. For example, cancer patients who were mid-course in chemotherapy lacked medical records, which are critical to administer the next round of treatment. Patients with prescriptions lacked documentation, and therefore could not readily get refills. Doctors at Emory found themselves recreating patient records based only on patient accounts. In addition, the hospital administration encountered financial, legal, and social service issues as a result of the imported surge. 21
These glimpses from actual experiences merely highlight how extreme weather can cause building failures that ultimately lead to interruptions in the continuity of care. The impacts are not momentary. From PTSD to the costs of structural rebuilding, the affects can be long-lasting. Yet, structural mitigation, while not a full panacea, can offer some protection to keep buildings and people safe.
The World Health Organization (WHO), which launched its “Save Lives. Make hospitals safe in emergencies” campaign in 2009 to raise attention to the number of (health and societal) dominoes that fall when disasters strike hospitals, says that the “most costly hospital is the one that fails.” This is because prevention, in the form of mitigation, is much less than the direct cost of repair and indirect cost of rebuilding the community around it.
WHO calculates that the price for retrofitting the non-structural items (e.g., moveable carts, trays) costs as little as 1% of the value of a hospital, while possibly protecting up to 90% of the hospital’s assets. According to FEMA, the most common points of hospital failure from storms are the elevator crankcases, windows and generators. Bolstering protection of these building assets costs magnitudes less than the cost to rebuild. Because of the expensive equipment, hospital damages range from $600,000 to $ 2 billion per facility. On the other hand, retrofitting an elevator crankcase, to maintain vertical movement is on the order of $300,000, and placing protective cover over the windows or moving the generators are likewise far less than the cost of rebuilding. Thus, mitigation for hospital buildings is likened to the health adage that says “an ounce of prevention is better than a pound of cure.”
Linking science for the protection of society is a matter of science-policy. It brings knowledge from peer-reviewed science journals into the public arena for consideration and application. Specifically, it involves combining the know-how from meteorology, the structural fortification from engineering, and the guidance and goal-setting from policy. This approach brings together all needed disciplines because no single field bears a monopoly on the knowledge that is needed to make a difference. With the increase in weather-related hazards 22 globally (and in the United States), and the irreplaceable importance of hospitals to society, both before, during, and after disasters, there is a need to focus on how to protect the healthcare infrastructure layer and to increase the chances that society can weather the storms yet to come.
Wendy Marie Thomas is a meteorologist for the AMS Policy Program. She researches science-policy issues relating to climate and weather scale impacts to human health and health care buildings.
References Schmidt, S., C. Kemfert, and P. Höppe, “Tropical cyclone losses in the USA and the impact of climate change: A trend analysis based on data from a new approach to adjusting storm losses,” Environmental Impact Assessment Review, vol 29, pp. 359-369, 2009.
 Höppe, P., and T. Grimm, “Rising natural catastrophe losses: What is the role of climate change?” Economics and Management of Climate Change: Risks, Mitigation and Adaptation, B. Hansjürgens and R. Antes, eds., Springer, pp. 13-22, 2009.
 Bruce, J.P., “Disaster loss mitigation as an adaptation to climate variability and change.” Mitigation Adaptation Strategies Global Change, vol. 4, pp 295-308, 2009.
Pielke, R., Jr., S. Agrawala, L Bouwer, “Clarifying the attribution of recent disaster losses: A response to Epstein and McCarthy,” Bulletin of the American Meteorological Society (BAMS), vol. 86, pp. 1481-1483, October 2005.
 Changnon, S.A., R.A. Pielke, Jr, D. Changnon, R.T. Sylves, and R. Pulwarty, “Human factors explain the increased losses from weather and climate extremes,” Bulletin of the American Meteorological Society (BAMS), vol. 81, pp. 437-482, March 2000.
 National Weather Service, “Tornadoes, Nature’s Most Violent Storms”, National Oceanic and Atmospheric Administration.
 Ocean and Coastal Resource Management, “The Coastal Community Development Partnership,” National Oceanic and Atmospheric Organization (NOAA), June 28, 2010. March 15, 2010,
 Office of Climate and Weather Services, “Weather Fatalities”, National Oceanic and Atmospheric Administration (NOAA), http://www.weather.gov/om/hazstats.shtml, based on preliminary data published: May 3, 2011.
 Federal Emergency Management Agency (FEMA), “Forward and Acknowledgements”, Design Guide for Improving Hospital Safety in Earthquakes, Floods, and High Winds: Providing Protection to Buildings and People (product #577), page: i, June 2007.
 Wolchover, Natalie, “Joplin, Missouri: Why are there no tornado building codes in Tornado Alley?” Christian Science Monitor: May 23, 2011.
 Goldstein, Amy, U.S. health-care expenditures up only 4 percent in 2009, suggesting effects of recession, Washington Post, January 5, 2011.
 Eaton, Leslie: New Orleans Recovery is Slowed by Closed Hospitals, New York Times, , July 24, 2007.
 National Weather Service Office for Springfield, MO, “Joplin Tornado Event Summary,” National Oceanic and Atmospheric Administration,. May 22, 2011.
 CNN Wire Staff, “Damaged Joplin hospital almost the only building left standing in area,” www.cnn.com, May 23, 2011.
 National Weather Service Forecast Office for Peachtree, GA, “Tornadoes in Central Georgia,” National Oceanic and Atmospheric Administration, March 1, 2007,.
 Susie Fussell, Chief RN Sumter Regional Hospital, presentation at the AMS’ “Rising Above the Weather” Forum, April 2009.
 Steve Storbakken, Director of Safety and Emergency Preparedness for Providence San Fernando Valley Area, presentation at the AMS’ “Rising Above the Weather” Forum, April 2009.
 National Climatic Data Center, 2008 Midwestern U.S. Floods, National Oceanic and Atmospheric Administration,.
 Chad Ware, Nurse and Program Coordinator at Mercy Medical Center, presentation at the AMS’ “Rising Above the Weather” Forum, April 2009.
 United States Congress: “A Failure of Initiative: Final Report of the Select Bipartisan Committee to Investigate the Preparation for and Response to Hurricane Katrina”, Washington, DC, Government Printing Office, February 19, 2006.
 American Meteorological Society, Policy Program Under the Weather Report, pp.4-6, October 2010.
 Munich, Re: “TopicsGeo natural catastrophes 2009: Analyses, assessments, positions.” Munich Reinsurance Company Rep., 40 pp., 2010.