Childhood drowning in Georgia: A geographic information system analysis
Drowning is one of the major causes of childhood mortality from unintentional injury in the United States. This study aims to delineate the drowning hot spots by aquatic environment and to examine the socioeconomic characteristics of neighborhoods with high densities. Drowning death records (2002-2008) among children aged 0-17 were obtained from the Office of the Child Advocate in Georgia. Drowning victims are initially analyzed by race, gender, age, aquatic environment, use of a personal floatation device, and the security of a pool. Using kernel density estimation and LISA statistics, this paper delineates drowning clusters by density and rate respectively. High-density neighborhoods are then investigated using a variety of socioeconomic factors by multiple-comparison test, Ordinary Least Squares (OLS) regression, and Geographically Weighted Regression (GWR). The result shows that the majority of victims were white, male, and children aged 0-4. Private swimming pools and open water were two leading drowning places. Blacks were more likely to drown in public swimming pools and in children aged 5-14 than white counterparts. Private swimming pools and children aged 0-4 have seen more white victims than blacks. Substantial spatial variation in drowning density and rate is present in Georgia. High-density drowning neighborhoods had a significantly lower rank of median income (p < 0.01) and lower rank of educational attainment in their counties than the low-density group. The GWR results suggest such associations vary in space. This study highlights the role of different socioeconomic factors in the interconnections of drowning risk factors. Findings suggest that health policies and prevention campaigns may be customized based on specific aquatic types and risk factors in different local communities. © 2012 Elsevier Ltd.
Dai, D; Zhang, Y; Lynch, CA; Miller, T; Shakir, M
Volume / Issue
Start / End Page
International Standard Serial Number (ISSN)
Digital Object Identifier (DOI)