A comparison of the geographic patterns of HIV prevalence and hurricane events in the United States

Abstract

Objectives: Previous research has documented the adverse association between hurricanes and HIV health outcomes; however, no research has been conducted to examine whether there is spatial overlap between hurricane events and the epidemiology of HIV in the United States. The objectives of this study were to assess the geographic distributions of and spatial correlation between HIV prevalence and the occurrences of hurricanes in the U.S. Study design: Cross-sectional study. Methods: Data on HIV prevalence rates were obtained from the Centers for Disease Control and Prevention, and data on U.S. hurricane events between 1851-2017 were obtained from the National Oceanic and Atmospheric Administration. ArcGIS v10.6 was used for mapping HIV prevalence rates and occurrences of hurricane events, and GeoDa v1.12.1.131 was used to conduct all univariate and bivariate spatial autocorrelation analyses. Results: HIV and hurricanes primarily affected states located in the south and along the Gulf Coast. Major hurricanes were reported among these regions also. States recording 20 or more hurricanes between 1851-2017 had an average HIV prevalence rate of 447.0 cases per 100,000. States recording five or more major hurricanes between 1851-2017 had an average HIV prevalence rate of 416.4 cases per 100,000. Regarding univariate spatial autocorrelation, HIV prevalence was clustered (Moran's I: 0.1825; P: 0.002). Hurricane events were also clustered (Moran's I: 0.2792; P: 0.006), as were major hurricanes (Moran's I: 0.1961; P: 0.018). There was no statistically significant bivariate spatial autocorrelation between HIV and hurricanes nor HIV and major hurricanes.