Hurricane Georges struck the Carribean Islands in September 1998, causing numerous deaths and extensive damage throughout the region. The Dominican Republic was hardest hit, with approximately 300 deaths; extensive infrastructure damage; and severe agricultural losses, including staple crops of rice, plantain, and cassava. Two months after the hurricane, the American Red Cross (ARC) was asked to provide food to an estimated 170,000 families affected by the storm throughout the country. To assist in directing relief efforts, CDC performed a needs assessment to estimate the food and water availability, sanitation, and medical needs of the hurricane-affected population. This report summarizes the results of that assessment, which indicate that, 2 months after the disaster, 40% of selected families had insufficient food ≥5 days per week and 28% of families reported someone in need of medical attention.
A household survey was performed using a modified cluster-sampling method1 to select persons from the first 33,000 families identified by the ARC as beneficiaries to receive support. The country was divided geographically into clusters. Using a random number generator, 30 clusters were selected with probability proportional to the number of households within a cluster. One adult family member was interviewed from each of seven selected households within each cluster. A total of 207 interviews were completed, representing 1414 persons. Respondents were asked about availability of food, water, and housing, and medical needs and storm preparation.
Data were analyzed using EpiInfo 6.1.2 Frequencies of variables were calculated for the population as a whole. Two groups of beneficiaries were considered by ARC: persons residing in groups in migrant farm worker settlements (i.e., bateys) (35 of 207 households) where baseline conditions were thought to be more harsh, and persons residing in shelters (50 of 207 households) who had been displaced from their homes and resources. Conditions for persons in bateys and shelters were compared with conditions for persons residing in other housing at the time of the survey.
The availability of food decreased dramatically after the hurricane. Respondents from 167 (83%) of 202 households reported insufficient food since the hurricane, compared with 107 (53%) of 202 who reported insufficient food before the storm (p<0.01). Of 202 households, 174 (86%) reported not having enough food ≥1 day per week. Of 201 households, 13 (7%) reported insufficient food for ≥5 days per week before the storm and 59 (29%) of 203 households after the storm. Of 202 households, 70 (35%) reported having enough food in their home for the next 3 days. At the time of the survey, relief services were providing food for 45 (23%) of 207 households; 38 (82%) of these families reported insufficient food despite the support.
Persons residing in batey households reported a higher mean number of days each week with insufficient food than did persons residing in other households before the hurricane (2.7 days compared with 1.9 days) (p<0.01). At the time of the survey, no difference was reported (3.4 days compared with 3.3 days). Of 33 batey households, 10 (30%) reported insufficient food ≥5 days per week, compared with 35 (24%) in the other households (p>0.05). Three fourths of batey households reported needing food; 16 (50%) batey households were relying on relief efforts for food.
Before the hurricane, the need for food was similar between households now housed in shelters and other nonbatey households. After the hurricane, 70% of nonbatey households identified food as a need, with 20 (40%) of 50 shelter households without enough food ≥5 days per week. At the time of the survey, relief efforts provided food for 32% of families in shelters.
At the time of the survey, 47 (28%) of 171 households reported having someone in the home who needed medical attention, and 160 (78%) of 205 households reported someone in the home who needed medication. Since the hurricane, 168 (82%) of 206 households reported an illness in a household member: respiratory illness (99 [59%]), gastrointestinal illness (69 [41%]), chronic illness (30 [18%]), or stress reaction (15 [9%]). In bateys and shelter populations, more families reported gastrointestinal (52% and 54%, respectively) (p<0.01) and respiratory (67% and 66%, respectively) disease than other families (37% [p<0.01] and 56% [p>0.05], respectively), but both were less likely to request medication. Families residing in shelters had the same access to health care than other households (67% compared with 74%) (p=0.4) and a higher self-reported need for health care (48% compared with 20%) (p<0.01).
After the storm, 97 (47%) households had running water. At the time of the survey, 18 (9%) households were relying on river water and 18 (9%) on rainwater. Although 85 (41%) households reported having wells, many reported water as a need. Most (93%) households reported access to a bathroom or latrine.
Dominican Republic Red Cross; American Red Cross International Svcs. National Center for Environmental Health; and EIS officers, CDC.
After a natural disaster, assessments typically are conducted within the first week to 10 days to determine the acute impact on the population.3 This evaluation was conducted 2 months after the storm to identify public health needs that remained after the emergency response to the disaster. Basic subsistence and health-care needs were present 2 months after the hurricane, when relief efforts were decreasing. Because of small sample sizes, comparisons between groups (i.e., batey, shelter, and other populations) should be interpreted with caution.
Following this assessment, recommendations were given to ARC and the Dominican Republic Ministry of Health, emphasizing the need for food, with special consideration of pregnant and lactating women and their newborns. In addition, periodic reassessments were recommended to monitor the effectiveness of follow-up interventions.4 Because of the high medical needs and low food availability reported by the shelter families, immediate preventive interventions were recommended.
As a result of these recommendations, ARC's food delivery schedule was accelerated by 1 month because of acute food needs. In addition, the Ministry of Health initiated medical interventions as soon as possible for shelter residents. Periodic needs assessments have been scheduled through November 1999.
Needs Assessment Following Hurricane Georges—Dominican Republic, 1998. JAMA. 1999;281(10):890-891. doi:10.1001/jama.281.10.890