Murine typhus is a common flea-borne disease caused by the bacterium Rickettsia typhi. In the United States, it is found in California, Hawaii, and Texas. The disease is typically maintained in a cycle with rodents and their associated fleas. However, it has also been found to be maintained in a peridomestic cycle containing opossums, free-ranging cats (feral and outdoor cats), dogs, and their associated fleas in Texas and California. Currently, there is a lack of information on the spatial pattern of Murine typhus cases, environmental predictors of incidence, and spatial variation of environmental predictors. More broadly, there is a lack of information on the influences of environmental factors of flea-borne diseases. The intention of this work is to investigate the spatial distribution of cases using methods of global and local spatial autocorrelation. In addition, this work is intended to identify environmental predictors of incidence and how they may very over geographic space by fitting a generalized linear model and geographically weighted regression model. Analysis was conducted with county-level case data for 2008 -2017 provided by the Texas Department of State Health Services (TDSHS), census data from TDSHS, and environmental data from the PRISM Climate Group.
Results/Conclusions
A Poisson linear regression model was fit with annual precipitation, minimum annual temperature, maximum annual temperature, mean annual dew point, minimum vapor pressure deficit, and annual maximum pressure deficit as predictor variables. All variables except for average minimum temperature and minimum vapor pressure deficit showed to be significant predictors of murine typhus incidence. However, these environmental variables may vary spatially throughout Texas which can be assessed with geographically-weighted regression.