Modeling Potential Future Lyme Disease Emergence Patterns Based on Projected Land Cover Under Different IPCC Scenarios

The emergence of infectious diseases over the past several decades has highlighted the need for new approaches to better understand and prepare for epidemics as endemic infectious diseases expand their geographic range. Specifically, the range of Lyme disease, the most common vector-borne disease in the United States, is currently emerging southward along the East Coast. An infection with Borrelia burgdorferi, the bacterium that causes Lyme disease, creates significant public health burdens so an improved understanding of geographic and ecologic dimensions of the disease's future emergence pattern will enhance diagnosis and reporting efforts, and potentially decrease numbers of human cases.

Research indicates that land cover is the most important contributor to Lyme disease emergence, and while a handful of other studies have considered future Lyme disease distributions under projected climate conditions, this study is the first to forecast potential spatial distributions of Lyme disease based on projected land cover. Using a previously developed statistical model, we predict future distributions of Lyme disease in Virginia based on land cover conditions under two different IPCC scenarios (A1B and A2) from 2010 to 2060. A Poisson point process is implemented with land cover and demographic parameters as the spatial covariates. Risk maps for both IPCC scenarios were generated from 2010 through 2060 by estimating incidence in each county in Virginia, and the land area and population affected were calculated. Results show that the A1B scenario has a 22.34% higher incidence estimate for the year 2060 than the A2 scenario. In the A1B scenario, 28% of the total area of Virginia is under high risk in 2010 which increases to 66% in 2060; in the A2 scenario, 22.4% is under high risk in 2010 with an increase in 2060 to 62.7%. In terms of population, 22% of Virginia residents are in the high risk zone for A1B scenario in 2010, with an increase to 42.9% in 2060; in the A2 scenario, 21% of the population is under high risk in 2010 and increases to 33% in 2060. Observed Lyme disease incidence data was used for model validation. Our results suggest that Lyme disease will continue to emerge in Virginia, with variation based on specific future land cover conditions.