The impacts on human health associated with flood events can be direct, including risk of drowning and injury, or indirect, such as increased risk of food-, water-, and vector-borne diseases. Immediate causes of death during floods include drowning and injury, with drowning accounting for the majority of flood-related deaths. Drowning deaths may occur during the rainfall or storm event that produces the flood or after the event when water levels are higher and rivers and streams are flowing faster than normal. Drowning may also occur when vehicles are caught in floodwaters. Injury is also a significant cause of morbidity and mortality during and after a flood. During a flood event, risk of injury is related to a person or vehicle being caught in the floodwaters during evacuation attempts. After a flood, injury may occur as a result of clean-up activities.
To examine the health impacts of flooding on Florida residents from 2005 through 2015, we conducted an ecological analysis comparing rates of hospitalizations and emergency department (ED) visits for all-cause, unintentional injury during flood impact periods and non-flood control periods. We obtained flood event data from the National Centers for Environmental Information (NCEI) Storm Events Database, and included all events defined as floods, coastal floods, flash floods, and lakeshore floods. We obtained injury data from the Florida Agency for Health Care Administration, and included all visits considered injury-related, as defined by International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes listed in primary or secondary diagnoses fields. Visits were aggregated to the county level for analysis. Each county served as its own control, matching a flood event period (impact period) with a period of similar length in which a flood did not occur (control period). The impact period included the entire flood event (i.e., some flood events occurred over multiple days) plus two weeks to include post-flood clean-up activities. We used conditional Poisson regression models to compare rate in the impact vs. control periods.
There were 285 unique flood events impacting all of Florida’s 67 counties during the period of interest, for a total of 462 impacts (flood event times counties impacted). Flood impacts included coastal floods (4.8%), floods (18.4%), and flash floods (23.7%). No lakeshore floods were reported during the period. The most affected areas included northwest coastal areas and the Panhandle. For the initial analysis, we used two control periods for every impact period and combined all types of injury. No increased rates of hospitalization or ED visits were found. Sub-group analyses looking at specific types of injuries are being conducted, as well as adjustments for overlapping impact and control periods for counties reporting multiple flood events during the same season that were identified after these first analyses.
Florida is vulnerable to frequent flooding and its associated impacts; however, the nature of this hazard varies. While initial analyses showed no difference in rates of ED visits and hospitalizations during impact and control periods, we expect there to be higher rates of specific types of injuries during impact periods in subsequent analyses. Such information will be key to planning public health prevention messages during and after major flooding events, and will assist in resource planning and allocation during response activities.