Deaths Related to Hurricane Irma — Florida, Georgia, and North Carolina, September 4–October 10, 2017
Weekly / August 3, 2018 / 67(30);829–832
Anindita Issa, MD1,2; Kirtana Ramadugu, MPH3,4; Prakash Mulay, MBBS3; Janet Hamilton, MPH3; Vivi Siegel, MPH2; Chris Harrison, MPH5; Christine Mullarkey Campbell6; Carina Blackmore, DVM, PhD3; Tesfaye Bayleyegn, MD2; Tegan Boehmer, PhD2 (View author affiliations)
View suggested citationThree powerful and devastating hurricanes from the 2017 Atlantic hurricane season (Harvey [August 17–September 1], Irma [August 30–September 13], and Maria [September 16–October 2]) resulted in the deaths of hundreds of persons. Disaster-related mortality surveillance is critical to an emergency response because it provides government and public health officials with information about the scope of the disaster and topics for prevention messaging. CDC’s Emergency Operations Center collaborated with state health departments in Florida, Georgia, and North Carolina to collect and analyze Hurricane Irma–related mortality data to understand the main circumstances of death. The most common circumstance-of-death categories were exacerbation of existing medical conditions and power outage. Further analysis revealed two unique subcategories of heat-related and oxygen-dependent deaths in which power outage contributed to exacerbation of an existing medical condition. Understanding the need for subcategorization of disaster-related circumstances of death and the possibility of overlapping categories can help public health practitioners derive more effective public health interventions to prevent deaths in future disasters.
Hurricane Irma, a Category 5 hurricane (185-mph winds), caused catastrophic damage in the Caribbean before moving northwest and making landfall in Florida on September 10, 2017, as a Category 4 hurricane. Wind damage compromised power lines, and a storm surge caused extensive flooding, primarily along the coast. Irma affected the entire state of Florida; 7 million residents were evacuated (1) and 6.7 million utility customers lost power (2). As Irma traveled inland, it weakened to a tropical depression; despite this weakening, 75,000 customers in North Carolina (3) and >900,000 customers in Georgia (4) experienced storm-related power losses.
As part of CDC’s public health response to the hurricanes, the Epidemiology and Surveillance Task Force in the Emergency Operations Center tracked online media reports of hurricane-associated deaths and contacted states for confirmation (5,6). The Georgia Department of Public Health and North Carolina Department of Health and Human Services provided CDC with information on confirmed hurricane-related deaths. Concurrently, the Florida Department of Health identified deaths associated with Hurricane Irma through examination of vital statistics death data from the electronic death registration system, reports from the Florida Medical Examiners Commission, and media reports. To identify hurricane-related deaths, Florida used text-parsing algorithms to query “How Injury Occurred” and “Literal Cause of Death” fields on the death certificates. Researchers developed a circumstance-of-death categorization scheme based on previous research (7) and used it to classify Hurricane Irma–related deaths (Box). This report summarizes the circumstances of confirmed Hurricane Irma–related deaths from September 4 to October 10, 2017, in Florida, Georgia, and North Carolina and highlights the need for detailed analysis of disaster-related circumstances of death.
Among the 129 hurricane-related deaths identified in Florida, Georgia, and North Carolina, 123 (95.3%) occurred in Florida; 88 (68.2%) decedents were male, and the median age was 63 years (range = 1–99 years). Eleven (8.5%) deaths were directly related to the hurricane, 115 (89.1%) were indirectly related, and three (2.3%) were possibly related (Table).
The most common category of indirect circumstance of death was exacerbation of an existing medical condition (46; 35.7%) (Table). Specifically, 23 (17.8%) deaths were associated with chronic health problems, such as cardiac disease, that were exacerbated by stress and anxiety related to the hurricane. Three (2.3%) deaths in chronically ill patients were attributed to disruption of emergency medical services. The remaining 20 (15.5%) deaths associated with exacerbation of an existing medical condition could also be categorized as power outage–related deaths (Figure). Seventeen (13.2%) heat-related deaths were associated with lack of air conditioning, and three (2.3%) deaths occurred in patients whose medical treatment (e.g., supplemental oxygen) was electricity-dependent. Fourteen (10.9%) of the heat-related deaths occurred among geriatric patients with existing chronic diseases who resided in an assisted-living facility in Florida that was without power for several days during a period of hot weather after the hurricane’s landfall. An additional 27 (20.9%) power outage–related deaths were not related to exacerbation of an existing medical condition. These included 16 (12.4%) carbon monoxide poisonings.
Discussion
Currently no published standardized methodology exists for analyzing disaster-related mortality data to inform public health action and prevent additional deaths. The death certification reference guide released in October 2017 by the National Center for Health Statistics informs medical examiners and coroners about completing death certificates for disaster-related deaths (6). Public health practitioners can refer to this document to understand current disaster-related death certification processes and how they can collaborate with medical examiners and coroners to obtain the specific mortality data needed to shape disaster-related public health communication strategies. Many public health agencies use traditional surveillance systems and social media surveillance to collect near real-time morbidity and mortality data. The more accurate and thorough the information, the more specifically communicators can target vulnerable groups with appropriate prevention messages.
Because circumstance of death typically provides more detailed information than cause of death, using circumstance of death for disaster-related mortality data analysis is more likely to guide public health action. A single cause of death might be associated with multiple circumstances of death. For example, a cause of death such as “blunt force trauma” could be associated with a motor vehicle crash or being struck by a falling object. The specific circumstances can inform different prevention messages. However, abstracting circumstance of death from the death certificate is more challenging than ascertaining the cause of death. Whereas “Cause of Death” is a labeled field in the death certificate, circumstance of death is determined through assessment of information in other free-text fields within the death certificate, such as “How Injury Occurred.”
Literature on U.S. hurricane-related mortality from recent decades has categorized circumstances of death (8,9). However, because each disaster is unique, the level of detail available in the circumstance-of-death categories varies across these reports. Not all disasters require a detailed analysis of all death circumstances; however, subcategorization of the most prevalent circumstances of death might reveal additional information that can be used to inform public health messages and interventions. For example, the circumstance of exacerbation of an existing medical condition might include subcategories such as stress-induced, disruption of emergency medical services, and power outage.
Public health messaging about hurricane safety and prevention of hurricane-related injuries should be communicated effectively in the hurricane preparation and response phases (10). With a sound understanding that death circumstances can be subcategorized and that categories might overlap, public health practitioners can perform supplemental analyses that will inform more specific and effective public health messaging and interventions to reduce disaster-related injury and illness. By looking at overlapping circumstances of death, analysis of Hurricane Irma mortality data revealed two unique subcategories of heat-related and oxygen-dependent deaths in which power outage contributed to exacerbation of an existing medical condition. Deaths associated jointly with power outages and existing medical condition exacerbation can be minimized by prioritizing power restoration to locations with vulnerable populations, including elderly persons and those with chronic diseases who are especially prone to heat-related illness. In addition, public health messages emphasizing generator safety and widespread use of carbon monoxide detectors can help reduce power outage–related carbon monoxide poisoning.
The findings in this report are subject to at least one limitation. The data might not include all deaths related to Hurricane Irma. As during any disaster, delayed reports of indirectly related deaths might not be recorded because of the imposed end dates of disaster-related mortality surveillance. In addition, death certifiers might change and refine circumstances of death as new information becomes available after registration of the death certificate; these death records are not included. Understanding the need for subcategorization of disaster-related circumstances of death can help public health practitioners develop more effective public health interventions to prevent deaths in future disasters.
Acknowledgments
The 2017 Hurricane Response Incident Management System Team, CDC Emergency Operations Center.
Conflict of Interest
No conflicts of interest were reported.
Corresponding author: Tegan Boehmer, tboehmer@cdc.gov, 770-488-3714.
1Epidemic Intelligence Service, CDC; 2Division of Environmental Health Science and Practice, National Center for Environmental Health, CDC; 3Florida Department of Health; 4Applied Epidemiology Fellowship Program, Council of State and Territorial Epidemiologists; 5Georgia Department of Public Health; 6North Carolina Department of Health and Human Services.
References
- Breslin S. FEMA chief to Florida Keys residents: “you’re on your own.” The Weather Channel. September 9, 2017. https://weather.com/storms/hurricane/news/hurricane-irma-florida-tampa-miami-impacts
- Florida Division of Emergency Management. Gov. Scott: more than 65 percent of power outages restored statewide. Tallahassee, FL: Florida Division of Emergency Management; 2017. https://www.floridadisaster.org/news-media/news/gov.-scott-more-than-65-percent-of-power-outages-restored-statewide/
- WRAL News. About 45,000 without power after Irma blows through parts of NC. WRAL News. September 12, 2017. https://www.wral.com/about-45-000-without-power-after-irma-blows-through-parts-of-nc/16942561/
- Chappell B. Power outages persist for millions in Florida, Georgia and Carolinas after Irma. NPR: The Two-Way. September 13, 2017. https://www.npr.org/sections/thetwo-way/2017/09/13/550674848/power-outages-persist-for-millions-in-florida-georgia-and-carolinas-after-irma
- CDC. Death scene investigation after natural disaster or other weather-related events toolkit: first edition. Atlanta, GA: US Department of Health and Human Services, CDC; 2017. https://www.cdc.gov/nceh/hsb/disaster/docs/DisasterDeathSceneToolkit508.pdf
- National Center for Health Statistics. A reference guide for certification of deaths in the event of a natural, human-induced, or chemical/radiological disaster. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics; 2017. https://www.cdc.gov/nchs/data/nvss/vsrg/vsrg01.pdf
- Combs DL, Quenemoen LE, Parrish RG, Davis JH. Assessing disaster-attributed mortality: development and application of a definition and classification matrix. Int J Epidemiol 1999;28:1124–9. CrossRef PubMed
- Jani AA, Fierro M, Kiser S, et al. Hurricane Isabel-related mortality—Virginia, 2003. J Public Health Manag Pract 2006;12:97–102. CrossRef PubMed
- CDC. Deaths associated with Hurricane Sandy—October–November 2012. MMWR Morb Mortal Wkly Rep 2013;62:393–7. PubMed
- CDC 2017 Hurricane Incident Management System Team. Hurricane season public health preparedness, response, and recovery guidance for health care providers, response and recovery workers, and affected communities—CDC, 2017. MMWR Morb Mortal Wkly Rep 2017;66:995–8. CrossRef PubMed
No hay comentarios:
Publicar un comentario