Urban disaster mitigation and preparedness: The 1999 Kocaeli earthquake (Turkey).

摘要

This study investigates the causes of deaths and injuries in the August 17th, 1999 Kocaeli earthquake to provide recommendations for risk mitigation. An epidemiological survey of the causes of deaths and injuries was completed for 453 households with 1,861 members. A geo-spatially stratified sample was selected randomly from residents of all 5 prefabricated cities and each of 5 neighborhoods in Central Golcu¨k 18--20 months after the disaster. The sample was representative of building damage levels in the area. Reinforced concrete buildings, buildings 5 floors and more, and buildings constructed after 1976 were all associated with more serious building damage and higher rates of deaths and injuries. The rate of death was 2% and injury 13.5%. Of the injured, 47% reported minor injuries, 45% moderate, and 8% severe; 12% were permanently disabled. All the deaths occurred in buildings damaged beyond repair. Injuries occurred in all damage states with 17% of moderate injuries occurring in less damaged buildings. Women were slightly more likely to be injured or killed than men, as were children ages 7--19. Adults ages 30--49 were more likely to be injured. Prior disability, older age, education level, income level were not significant factors in death and injury. Being in bed was safer than being awake or sitting. The association of deaths with injuries to head neck and chest, and wide range of injuries reinforces the objective to get down low, make yourself small and cover head and neck during shaking. The 16.5% of injuries sustained after the earthquake, and 35% of injuries to feet and legs suggest the importance of keeping shoes and flashlight by the bed to protect from nighttime injury. Striking, cutting, or pinning objects caused 85% of the injuries. Of these 32% were structural building elements, 31% non-structural building elements (especially infill walls) and 53% building contents. Of the deaths, 74% were due to structural object, 34% non-structural objects and 11% building contents. Empirical findings are supplemented by content analysis of Turkish and U.S. public education documents and description of research utilization activities.