The content on this page is being archived for historic and reference purposes only. The content, links, and pdfs are no longer maintained and might be outdated.
Morbidity and Mortality Associated With Hurricane Floyd --- North Carolina, September--October 1999
Please note: An erratum has been published for this article. To view the erratum, please click here.
On September 16, 1999, Hurricane Floyd, a storm extending 300 miles with sustained winds of 96--110 miles per hour, made landfall in North Carolina, dropping up to 20 inches of rain in eastern regions of the state. Rain from Hurricane Floyd, combined with rains from Hurricane Dennis beginning on August 30 and Hurricane Irene on October 17, caused extensive flooding along the Neuse, Tar, Roanoke, Lumbar, and Cape Fear rivers, affecting an estimated 2.1 million persons. This report presents data about injuries, illnesses, and deaths during and following Hurricane Floyd in North Carolina and identifies the leading cause of death as drowning involving occupants of motor vehicles trapped in flood waters.
Epidemiologic information about deaths related to Hurricane Floyd were provided to CDC by the state medical examiner's office. To monitor illness and injury related to the hurricane and subsequent flood, emergency department (ED) surveillance was established at 20 hospitals in 18 flood-affected counties in eastern North Carolina. Standardized illness and injury classifications were developed and applied by a disaster response team and ED staff during the surveillance period for comparison with similar periods in 1998. Diagnosis or chief symptoms for each patient visit was abstracted from daily ED logs to monitor trends during September 16--October 27, 1999. The 1999 illness and injury data were compared with data from 4 days in September 1998 (September 13 [Sunday], 15 [Tuesday], 17 [Thursday], and 19 [Saturday]) and 4 days in October 1998 (October 11 [Sunday], 13 [Tuesday], 15 [Thursday], and 17 [Saturday]). To compare a complete week of 1998 data with 1999 data, the September 1998 weekdays were weighted by multiplying by 2.5 and added to the weekend days; the same methods were applied to October 1998 data. Analysis of variance was used to compare the number of ED visits for each weekday during the 1999 surveillance period.
The medical examiner determined that 52 deaths were associated directly with the storm. Decedents ranged in age from 1 to 96 years (median: 43 years); 38 (73%) were males. Twenty counties reported at least one death; 40% of all deaths occurred in three counties. Of the 52 deaths, 35 (67%) occurred on September 16. The leading cause of death was drowning (Table 1); 24 (67%) deaths involved occupants of motor vehicles trapped in flood waters. Seven deaths occurred during transport by boat; flotation devices were not worn by any of the decedents. Five (10%) of the 52 decedents were rescue workers.
During September 16--October 27, 59,398 ED visits were reported; 67% related to illnesses and 33% to injuries. Four conditions accounted for 63% of all visits: orthopedic and soft tissue injury (28%), respiratory illness (15%), gastrointestinal illness (11%), and cardiovascular disease (9%); 19 cases of hypothermia occurred following the hurricane, including one death. EDs reported no hypothermia cases during the 1998 reference period. During the 1999 surveillance period, 10 cases of carbon monoxide poisoning were reported, compared with none during the 1998 reference period.
No statistical differences were found when comparing the number of ED visits with different days of the week during the surveillance period in 1999. Comparing the first week following Hurricane Floyd with the first week of September 1998, significant increases were reported in suicide attempts (relative risk [RR]=5.0; 95% confidence interval [CI]=1.4--17.1), dog bites (RR=4.1; 95% CI=2.0--8.1), febrile illnesses (RR=1.5; 95% CI=1.3--1.9), basic medical needs (e.g., oxygen, medication refills, dialysis, and vaccines) (RR=1.4; 95% CI=1.2--1.8), and dermatitis (RR=1.4; 95% CI=1.2--1.6). Comparing a week 1 month after Hurricane Floyd with the same period in 1998, significant increases were reported in 1999 for arthropod bites (RR=2.2; 95% CI=1.4--3.4), diarrhea (RR=2.0; 95% CI=1.4--2.8), violence (i.e., assault, gunshot wounds, and rape) (RR=1.5; 95% CI=1.1--2.2), and asthma (RR=1.4; 95% CI=1.2--1.7). Routine surveillance by local public health workers following Hurricane Floyd identified outbreaks in shelters of self-limiting gastrointestinal disease and respiratory disease.
Reported by: S Beaman, Columbia Heritage Hospital, Tarboro; C Boone, Nash General Hospital, Rocky Mount; S Bowman, Carteret General Hospital, Morehead City; K Brown, Onslow Memorial Hospital, Jacksonville; J Burke, New Hanover Regional Hospital, Wilmington; C Davis, Wayne Memorial Hospital, Goldsboro; A Eason, Roanoke-Chowan Hospital, Ahoskie; P Etheridge, Albemarle Hospital, Elizabeth City; L Evans, Camp Legeune Naval Hospital, Jacksonville; L Fulcher, Beaufort County Hospital, Washington; H Jones, Halifax Memorial Hospital, Roanoke Rapids; A McDaniel, Lenoir Hospital, Kinston; A Monday, Columbia Brunswick Hospital, Supply; C Ohl, MD, D Hayes, MD, W Weist, MD, J Dolzinger, MD, Pitt Memorial Hospital, Greenville; C Peah, Pender County Hospital, Burgaw; C Shay, Dosher County Hospital, Southport; S Smith, Bertie Hospital, Windsor; A Thomas, Duplin General Hospital, Kenansville; C Warren, Wilson Memorial Hospital, Wilson; L Wheaton, Craven Regional Medical Center, New Bern; CJ Butts, MD, S Cline, DDS, D Enright, E Howell, D McBride, MD, J Reddington, J Wilson, E Zeringue, N MacCormack, MD, State Epidemiologist, North Carolina Dept of Health and Human Svcs. Immunization Svcs Div, National Immunization Program; Div of TB Elimination, National Center for HIV, STD, and TB Prevention; Health Studies Br, Div of Environmental Hazards and Health Effects, National Center of Environmental Health; State Br, Div of Applied Public Health Training, Epidemiology Program Office; and EIS officers, CDC.
In areas where flash flooding occurs, water rises quickly, forcing persons to evacuate without preparation. During and after Hurricane Floyd, rural inland counties were the most severely affected (S. Yount, Federal Emergency Management Agency, personal communication, 2000). Persons residing in affected areas may not have recognized or been informed about the risks associated with severe storms. Most mortality and morbidity caused by inland hurricanes have been attributed to the effects of high winds (1--3); however, surveillance during and after Hurricane Floyd showed morbidity and mortality patterns similar to other flood-related disasters (4--6). Drowning was a major cause of death, especially among persons who attempted to drive through moving water.
Hurricane Floyd surveillance reports of nonfatal injuries and illnesses were similar to earlier storms, with reported increases in insect stings (2,7,8), dermatitis, diarrhea (8), and psychiatric conditions (9). Findings unique to Hurricane Floyd included increases in reports of hypothermia, dog bites, and asthma.
The findings in this report are subject to at least three limitations. First, the surveillance system was limited because the EDs did not represent the range of health-care services used by persons in flood-affected areas. Second, if ED logs contained misclassified diagnoses, some medical conditions might not have been identified and recorded properly. Third, on the basis of the assumption that diagnoses on weekdays do not vary, only 8 days of data were collected for September and October 1998, potentially limiting the strength of the comparison with 1999.
In the aftermath of Hurricane Floyd, some surveillance data suggest that public health intervention strategies could improve in future hurricane-related disasters. State agencies need to identify regional and local organizations that represent communities at risk. A coordinated disaster response could strengthen available resources and improve response scope and efficiency. Surveillance data also suggest that deaths from floods may be prevented by identifying flood-prone areas and advising persons at risk to take appropriate actions. Public service announcements, educational materials, and training programs on hurricane preparedness should be made accessible to all communities before the hurricane season. For example, motorists should be warned not to drive through areas in imminent danger of flash floods or onto roads and bridges covered by rapidly moving water. If vehicles are necessary to evacuate a community, safe evacuation routes should be identified in advance. In addition, all persons using boats for transport should wear flotation devices. The deaths of five rescue workers suggest the need for occupational risk prevention training. Persons should take precautions against dog bites and hypothermia (10), and persons with asthma returning to flooded homes should guard against exposure to mold and mildew that may exacerbate respiratory symptoms (10). Throughout all phases of disaster relief, appropriate mental health services should be made available. In anticipation of the August--November hurricane season, community disaster planning should begin by early spring.
Return to top.
Disclaimer All MMWR HTML versions of articles are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the electronic PDF version and/or the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.**Questions or messages regarding errors in formatting should be addressed to email@example.com.
Page converted: 5/4/2000
This page last reviewed 5/2/01