Enhanced spatial models for predicting the geographic distributions of tick-borne pathogens

Area of Science:

• Epidemiology
• Spatial Ecology
• Public Health

Background:

• Disease mapping is crucial for public health, yet emerging infectious disease data is often sparse.
• Tick-borne pathogens, such as Ehrlichia chaffeensis and Anaplasma phagocytophilum, pose significant health risks.
• Predicting geographic disease distributions requires advanced spatial modeling techniques.

Purpose of the Study:

• To compare spatial modeling approaches for predicting the geographic distributions of two tick-borne pathogens.
• To evaluate the impact of spatial autocorrelation and spatial heterogeneity on predictive accuracy.
• To enhance environmental modeling for infectious disease risk assessment.

Main Methods:

• Extended logistic regression models to include spatial autocorrelation and spatial heterogeneity.
• Applied models to predict distributions of Ehrlichia chaffeensis and Anaplasma phagocytophilum.
• Assessed model accuracy based on pathogen distribution characteristics.

Main Results:

• Models incorporating spatial autocorrelation or heterogeneity significantly outperformed standard logistic regression.
• For Ehrlichia chaffeensis, a model with spatial autocorrelation was most accurate due to its clustered distribution.
• For Anaplasma phagocytophilum, a model with both spatial autocorrelation and heterogeneity yielded the best predictions.

Conclusions:

• Spatial autocorrelation improves disease risk model accuracy by accounting for unmeasured spatial factors.
• Spatial heterogeneity enhances prediction by considering region-specific ecological influences on disease drivers.
• These advanced spatial methods are vital for accurate infectious disease mapping and risk assessment.