full-text ►
Deaths Associated with Hurricane Sandy — October–November 2012

	MMWR Weekly Volume 62, No. 20 May 24, 2013

Deaths Associated with Hurricane Sandy — October–November 2012

Weekly

May 24, 2013 / 62(20);393-397

On October 29, 2012, Hurricane Sandy* hit the northeastern U.S. coastline. Sandy's tropical storm winds stretched over 900 miles (1,440 km), causing storm surges and destruction over a larger area than that affected by hurricanes with more intensity but narrower paths. Based on storm surge predictions, mandatory evacuations were ordered on October 28, including for New York City's Evacuation Zone A, the coastal zone at risk for flooding from any hurricane (1). By October 31, the region had 6–12 inches (15–30 cm) of precipitation, 7–8 million customers without power, approximately 20,000 persons in shelters, and news reports of numerous fatalities (Robert Neurath, CDC, personal communication, 2013). To characterize deaths related to Sandy, CDC analyzed data on 117 hurricane-related deaths captured by American Red Cross (Red Cross) mortality tracking during October 28–November 30, 2012. This report describes the results of that analysis, which found drowning was the most common cause of death related to Sandy, and 45% of drowning deaths occurred in flooded homes in Evacuation Zone A. Drowning is a leading cause of hurricane death but is preventable with advance warning systems and evacuation plans. Emergency plans should ensure that persons receive and comprehend evacuation messages and have the necessary resources to comply with them.
Red Cross tracks deaths during disasters to provide services to surviving family members, including crisis counseling, assistance with disaster-related expenses, locating emergency housing, identifying recovery resources, and addressing disaster-related health needs. Red Cross volunteers search for reports of disaster-related deaths from sources such as funeral home directors, the Federal Emergency Management Agency (FEMA), hospitals, and news reports. Volunteers then obtain information about these deaths from sources including the medical examiner/coroner, physician, fire department/police, and family of the decedent (2).
Deaths included in this analysis were any Sandy-related death recorded on a Red Cross mortality form with a date of death up to November 30, 2012. Mortality forms included the decedent's age, sex, race (white, black, Asian, other, or unknown), and date and location of death. Disaster-related deaths were categorized as direct or indirect. Directly related deaths are deaths caused by the environmental force of the disaster (e.g., wind or flood) or by the direct consequences of these forces (e.g., structural collapse). Indirectly related deaths are defined as deaths occurring in a situation in which the disaster led to unsafe conditions (e.g., hazardous roads) or caused a loss or disruption of usual services that contributed to the death (e.g., loss of electrical services) (2). Deaths without direct or indirect classification were reported as unknown or possibly related deaths. Daily counts of direct, indirect, and unknown/possibly related deaths were calculated based on the dates of each death. The characteristics of drowning deaths were compared with all deaths using chi-square tests of trend and t-tests. Home addresses of decedents whose drowning death occurred in the home were examined with respect to FEMA's hurricane storm surge area (field-verified as of November 11, 2012 [3]) and known, geographically defined areas under evacuation order (i.e., New York City's Evacuation Zone A) (1).
A total of 117 deaths were reported on Red Cross mortality forms. The source of information for the mortality forms was a medical examiner/coroner for 94 (80.3%) cases and the family of the decedent for 10 (8.5%) cases (Table). Most deaths occurred in New York (53 [45.3%]) and New Jersey (34 [29.1%]); the other deaths occurred in Pennsylvania, West Virginia, Connecticut, and Maryland. The deaths occurred during October 28–November 29, 2012 (Figure 1). Approximately half of the deaths (60 [51.3%]) occurred on the first 2 days of the storm's landfall, with a peak of 37 deaths on October 30, 2012.
Decedents ranged in age from 1 to 94 years (mean: 60 years, median: 65 years); 60.7% were male, and 53.8% were white. Of the 117 deaths, 67 (57.3%) were classified as directly related deaths, and 38 (32.5%) were indirectly related to the storm. Of the directly related deaths, the most common mechanism was drowning (40 [59.7%]), followed by trauma from being crushed, cut, or struck (19 [28.4%]). Poisoning was the most common indirectly related cause of death; of the 10 poisonings, nine were caused by carbon monoxide. Most directly related deaths occurred during the first few days of the storm, whereas indirectly related deaths continued from the day before the storm into the middle of November.
Comparing the 40 drowning deaths to all Sandy-related deaths, the age, sex, and race distributions of decedents were similar (Table). The majority of drowning deaths (29 [72.5%]) also occurred in the initial phase of the storm, during October 29–31. Twenty-one (52.5%) drowning deaths occurred in the decedent's home, and 11 (27.5%) occurred outside; one person drowned in a flooded commercial building lobby, and another person drowned while intentionally swimming off a storm-affected beach. For six deaths, circumstances of the drowning were not available. The location of drowning deaths by state was significantly different (p<0 .05="" 18="" 27.3="" a="" accounted="" addresses="" all="" and="" class="callout-pink" compared="" deaths.="" deaths="" decedents="" drowned="" drowning="" evacuation="" flooded="" for="" home="" homes="" in="" located="" majority="" new="" nondrowning="" occurred="" of="" only="" sandy-related="" span="" the="" them="" twenty="" were="" whereas="" with="" york="" zone="">Figure 2
); the other two decedents' homes were in or near areas of flooding and near Evacuation Zone A. Notes written by Red Cross volunteers on these 20 deaths captured decedents' reasons for not evacuating, such as "afraid of looters," "thought Hurricane Irene was mild," and "unable to leave because did not have transportation."

Reported by

Mary Casey-Lockyer, MHS, Rebecca J. Heick, PhD, American Red Cross. Caitlin E. Mertzlufft, MPH, Div of Toxicology and Human Health Sciences, Agency for Toxic Substances and Disease Registry. Ellen E. Yard, PhD, Amy F. Wolkin, MSPH, Rebecca S. Noe, MPH, Div of Environmental Hazards and Health Effects, National Center for Environmental Health; Michelle Murti, MD, EIS Officer, CDC. Corresponding contributor: Michelle Murti, mmurti@cdc.gov, 770-488-3418.

Editorial Note

The "perfect storm" weather conditions of Hurricane Sandy resulted in extensive damage to infrastructure and large flood zones (4). The direct and indirect impacts of the storm led to challenging, and sometimes deadly, conditions for residents, including prolonged power outages, storm surges, and disrupted services. More than half (51.3%) of deaths from Sandy occurred within the first 2 days of the storm, and the most common cause of death was drowning. Approximately half of the drowning deaths were in flooded homes located in areas that were under mandatory evacuation orders as of October 28, 2012, the day before Sandy's landfall (1).
Before the 1970s, drowning from wind-driven storm surges was by far the most common cause of hurricane-related death (5). Advances in hurricane warning and evacuation systems have helped to reduce drowning deaths. Since that time, hurricanes have had other leading causes of death, such as trauma for the Florida hurricanes in 2004 and 2005, and carbon monoxide poisoning for Hurricane Ike in 2008 (6,7). However, drowning continues to be an important cause of death, and was the leading cause for Hurricane Katrina (2005) and Sandy (8).
The findings in this report are subject to at least two limitations. First, the number of deaths reported is limited to those captured through Red Cross mortality tracking, which is only activated in areas with a Red Cross Disaster Relief Operation. In an evaluation of Red Cross mortality tracking versus Texas' active disaster-related mortality surveillance during Hurricane Ike, Red Cross had a sensitivity of 47% (Red Cross cases compared with Texas cases) and positive predictive value of 92% (Red Cross Ike cases compared with all Red Cross cases); thus, the cases presented in this report are likely to be actual cases but are unlikely to include all Sandy-related deaths (2). Media sources have reported 131 fatalities in the United States from the storm (9); Sandy mortality statistics, including death certificates, are pending official release. Second, the specific location of death was only available for decedents who died at home, limiting other geographic comparisons. Additionally, New York City's Evacuation Zones provided the only geographic data available for identifying areas of evacuation; however, 95% of all drowning deaths at home were in or near these areas.
Hurricane-related drowning deaths in evacuation zones are preventable. A successful evacuation depends on officials providing timely messaging to all affected persons, on persons receiving those messages, and on persons having the capacity, resources, and willingness to evacuate. The penetration of evacuation messages to decedents or their communities was not assessed in this report, but future research should evaluate the effectiveness of the hurricane evacuation orders. Given the inability and unwillingness of some residents to evacuate, additional research is needed to identify barriers and motivators for persons during an evacuation and the effectiveness of interventions designed to assist these persons.

Acknowledgments

American Red Cross Disaster Health Svcs staff and volunteers deployed for Sandy. Linda Quan, MD, Univ of Washington School of Medicine and American Red Cross Scientific Advisory Council. Dana Flanders, PhD, Emory Univ, Atlanta, GA. Barry Flanagan, PhD, Elaine Hallisey, MA, Div of Toxicology and Human Health Sciences, Agency for Toxic Substances and Disease Registry.

References

1. Saul MH. Parts of New York City evacuated for Hurricane Sandy. Wall Street Journal; 2012. Available at http://online.wsj.com/article/SB10001424052970203880704578084701930663668.html.
2. Farag NH, Rey A, Noe R, Bayleyegn T, Wood AD, Zane D. Evaluation of the American Red Cross disaster-related mortality surveillance system using Hurricane Ike data—Texas 2008. Disaster Med Public Health Prep 2012; December 7 [Epub ahead of print]
3. Federal Emergency Management Agency. FEMA MOTF-Hurricane Sandy impact analysis. Washington, DC: US Department of Homeland Security, Federal Emergency Management Agency; 2012. Available at http://fema.maps.arcgis.com/home/webmap/viewer.html?webmap=307dd522499d4a44a33d7296a5da5ea0.
4. Federal Emergency Management Agency. Hurricane Sandy: timeline; 2012. Washington, DC: US Department of Homeland Security, Federal Emergency Management Agency; 2012. Available at http://www.fema.gov/hurricane-sandy-timeline.
5. National Oceanic and Atmospheric Administration. Hurricanes and tropical meteorology. Washington, DC: US Department of Commerce, National Oceanic and Atmospheric Administration; 2000. Available at http://www.aoml.noaa.gov/general/WWW000/nhurr00.html#imp2.
6. Ragan P, Schulte J, Nelson SJ, Jones KT. Mortality surveillance: 2004 to 2005 Florida hurricane-related deaths. Am J Forensic Med Pathol 2008;29:148–53.
7. Zane DF, Bayleyegn TM, Hellsten J, et al. Tracking deaths related to Hurricane Ike, Texas, 2008. Disaster Med Public Health Prep 2011;5:23–8.
8. Brunkard J, Namulanda G, Ratard R. Hurricane Katrina deaths, Louisiana, 2005. Disaster Med Public Health Prep 2008;2:215–23.
9. Reuters. Factbox: storm Sandy blamed for at least 132 deaths in U.S., Canada; 2012. Reuters; 2012. Available at http://www.reuters.com/article/2012/11/16/us-storm-sandy-deaths-idusbre8af0zx20121116.
10. Fairchild AL, Colgrove J, Jones MM. The challenge of mandatory evacuation: providing for and deciding for. Health Aff (Millwood) 2006;25:958–67.

* Sandy evolved from a Category 3 hurricane in the Caribbean to an intense post-tropical cyclone before landfall in the United States.