RESUMO
INTRODUCTION: In disaster situations, timely surveillance systems that provide illness, injury, and mortality information to public health officials and hospitals are essential for planning and evaluating interventions. OBJECTIVES: To describe flood surveillance methodology, the impact of the event on hospitals, and the number of daily patient visits due to selected illnesses and injuries before, during, and after severe flooding in southeastern Louisiana in May 1995. METHODS: Survey of disaster-area hospitals regarding flood impact. Emergency department surveillance of injuries and illnesses for the week before, the two days during, and the week after the flood. RESULTS: There occurred an increase in the number of persons who drowned or were injured that presented to the moderately affected hospitals during the storm, but there was no increase in visits for gastroenteritis to any group of hospitals. Services were disrupted in more than half of hospitals. The severely affected hospitals had the least variation in the average number of daily visits. None of the drownings were reported by those hospitals that reported severe service disruption. CONCLUSIONS: Data should be collected from all hospitals in or near disaster areas, even if they were not directly affected by the disaster. Public education about the danger of drowning during flash flooding must be improved. The Louisiana experience emphasizes the need for a disaster-preparedness plan for rapid surveillance of illnesses and injuries.