1 table omitted
On October 24, 2005, Hurricane Wilma, the most intense hurricane (882 mb) ever recorded in the Atlantic Basin, made landfall on the southern tip of Florida.1 By landfall, Wilma had been downgraded from a Category 5 (i.e., winds of ≥156 mph) to a Category 3 hurricane but still contained winds of >110 mph. The storm moved slowly over the Florida Keys and south Florida, causing extensive wind and flood damage to homes and businesses. Approximately 3 million households were left without power, and thousands of residents were displaced to temporary shelters; 10 deaths were storm related.2 On October 27, the Florida Department of Health (FDOH) asked CDC and the North Carolina Division of Public Health (NCDPH) for assistance in performing a rapid needs assessment of communities most affected by the storm. On the basis of information from local public health officials, field assessment teams, and electric power companies, FDOH identified Hendry County, a rural county with a 2000 population of 36,210, as the most severely affected area. Two Hendry County communities, Montura Ranch Estates and Pioneer Plantation in the town of Clewiston (pop. 6,460), were of particular concern. According to the 2000 census, nearly 20% of Clewiston residents lived below the poverty level.3 The assessment determined that approximately one third of households also had been affected by at least one other hurricane that preceded Wilma during 2004 and 2005. More than half of the households surveyed lived in homes that were damaged but still habitable, and 10% of households in Montura Ranch Estates reported their homes as uninhabitable; approximately 73% of households had not received information about how to remain safe during clean-up activities. Results of the assessment were immediately provided to the Hendry County Emergency Operations Center, Hendry County Health Department, and FDOH for allocation of resources to help these communities recover from the hurricane.
During November 1-2, teams from FDOH, CDC, NCDPH, and North Carolina Public Health Regional Surveillance Teams (PHRST) traveled to Clewiston, on Lake Okeechobee in south-central Florida, to assess the needs of residents in the two communities 8 days after Hurricane Wilma. The local tax appraiser's office provided information regarding property parcels, which was used to obtain a sample population. Any parcel that did not contain at least one building structure was excluded from the survey. Samples of 166 of 1,222 parcels thought to have structures in Montura Ranch Estates and 140 of 345 parcels in Pioneer Plantation were randomly selected using statistical analysis software, and locations were mapped as described previously.4,5 Because more than 40% of Clewiston residents were Hispanic, bilingual teams were deployed, and all survey questions were asked in English or Spanish.
In Montura Ranch Estates, teams used handheld devices equipped with global positioning system technology to locate the selected parcels and personal digital assistants (PDAs) to record survey responses. In Pioneer Plantation, teams used paper maps of the area to locate selected parcels. Interviews were conducted using both paper and PDA versions of the questionnaire.
Interviewers in both areas made at least four attempts, at least 1 hour apart, to make contact with the selected households. In all cases, the reason for an unsuccessful interview attempt was recorded. When an interview could not be obtained at a selected address, teams proceeded to the next address on their list.
During November 1-2, teams approached the 166 parcels in Montura Ranch Estates and 140 parcels in Pioneer Plantation. Ninety-one interviews were completed in Montura Ranch Estates (overall response rate: 55%) and 74 in Pioneer Plantation (overall response rate: 53%), for a total of 165 households. In Montura Ranch Estates, 17 (10%) of the 166 parcels approached contained no identifiable housing structure. Thirty-seven (22%) contained a housing structure that was not occupied at the time of the interview attempt. Ten (6%) homes were inaccessible (e.g., because of flooding or debris), and nine (5%) were destroyed. Two (1%) of the parcels were occupied by households that declined to be interviewed. In Pioneer Plantation, 12 (9%) of the 140 parcels approached contained no identifiable housing structure, and 39 (28%) contained homes that were unoccupied. Eleven (8%) homes were inaccessible, and two (1%) were destroyed. Households at two (1%) parcels declined to be interviewed.
In both Hendry County communities, the average household size was three persons, both before and after Hurricane Wilma. In Montura Ranch Estates, 11 (12%) households had at least one occupant aged <2 years, and 32 (35%) had at least one occupant aged ≥65 years. In Pioneer Plantation, one household reported an occupant aged <2 years, and three (4%) households had at least one occupant aged ≥65 years.
In Montura Ranch Estates, 74 (81%) households lived in mobile homes, and 17 (19%) lived in single-family homes. Forty-seven (64%) Pioneer Plantation households lived in mobile homes and 26 (35%) in single-family homes. Thirty-five (38%) homes in Montura Ranch Estates and 34 (46%) homes in Pioneer Plantation had minimal or no damage. Eighty-six (52%) of the homes in both areas were considered damaged but habitable by the residents living there at the time of the interview. Nine (10%) homes in Montura Ranch Estates were considered to be uninhabitable, and one (1%) home in Pioneer Plantation was considered uninhabitable. Eighteen (20%) households in Montura Ranch Estates and 29 (39%) in Pioneer Plantation reported needing a tarpaulin to cover leaking roofs. At the time of the survey, more than 88% of households in both communities had basic utilities (i.e., running water, working toilet, and landline or cellular telephone service). Two (2%) households in Montura Ranch Estates and seven (9%) in Pioneer Plantation reported having a working carbon monoxide (CO) detector.
Eighty-two (90%) households in Montura Ranch Estates and 58 (80%) in Pioneer Plantation reported receiving disaster relief (e.g., food, water, ice, or shelter) in the 7 days before the interview. A total of 145 of 150 (97%) households in both communities reported having access to a 3-day supply of food, and 19 (13%) reported that at least one household member had been ill or sustained an injury since the hurricane. Six (4%) of 150 households reported that a household member had been unable to obtain needed medical care.
Forty-five (27%) households in the two communities reported having received information about safety (e.g., safe use of pressure cleaners to reduce the risk for CO poisoning during hurricane cleanup) during the week before the survey. Radio and television were the most common sources of this information. Other sources included word of mouth and flyers.
W Service, MSPH, North Carolina Div of Public Health; S Ramsey, K Henderson, DO, M Smith, PhD, North Carolina Public Health Regional Surveillance Teams. S Emmanuel, MPH, Florida Dept of Health. D Batts, MD, C Sanchez, MD, S Young, MS, MPH, E Azziz-Baumgartner, MD, M Patel, MD, M Miller, L Corrales, MPH, National Center for Environmental Health; J Schulte, DO, Career Epidemiology Field Officer; L Conklin, MD, EIS Officer, CDC.
Rural areas present unique challenges to assessing and fulfilling the needs of residents after disasters such as hurricanes. Although they often sustain more damage than urban areas, needs of rural communities can be more difficult to assess because of lesser population density. Residents in certain rural areas are more socieoeconomically disadvantaged and often have more needs than their urban counterparts; available recovery resources can be quickly exhausted.
The survey results indicate that public health risk factors were present in Hendry County 8 days after Hurricane Wilma struck the area. Property damage was substantial, with more than half of the homes, although habitable, sustaining damage from the storm. The majority of residents had used disaster relief aid services, but only 27% reported receiving information on clean-up safety. Only 5% of survey participants had a working CO detector in their homes, although one third of households had been affected by at least one previous hurricane in 2004-2005.
The findings in this report are subject to at least three limitations. First, residents of evacuated or destroyed homes were unavailable for inclusion in the assessment, which likely underestimated overall needs. Second, homes that might have been occupied but were inaccessible because of flooding or debris also were not included in the assessment. Finally, lack of updated census information and maps to accurately define the communities being assessed might have resulted in sampling errors.
On November 3, recommendations based on the findings in this report were presented to FDOH and the Hendry County Emergency Operations Center, which responded immediately. Recommendations included the following: (1) provide residents with information and assistance on post-storm home remediation (e.g., clean-up safety, debris removal, and CO poisoning risks), (2) restore electric power to households without it, (3) consider implementing a program to facilitate access to medical care and prescription drugs, (4) provide tarpaulins to residents with damaged roofs to temporarily prevent leaking, and (5) ensure that populations affected by the storm had knowledge of and access to food banks in their areas.
REFERENCES: 5 available
Rapid Needs Assessment of Two Rural Communities After Hurricane Wilma—Hendry County, Florida, November 1-2, 2005. JAMA. 2006;296(1):34-36. doi:10.1001/jama.296.1.34