You are here:
Forest fragmentation and Lyme disease
Citation:
Hilborn, E AND L. Jackson. Forest fragmentation and Lyme disease. Presented at Duke Conservation Society Spring Symposium, Durham, NC, March 28, 2014.
Impact/Purpose:
Our model of reported human Lyme disease and landcover characteristics suggest that a 50:50 mix of forest and herbaceous cover with maximum common boundaries (edge) between the classes is the highest risk landscape for acquisition of illness. These findings are consistent with the known biological phenomenon of Lyme disease transmission, where vectors may thrive in degraded habitats. Results suggest that land use planning may provide a viable tool to reduce the risk for human acquisition of Lyme disease.
Description:
Lyme disease is the most commonly reported vectorborne disease in the United States. It is associated with human exposure to infected Ixodes ticks which exist even in degraded forest and herbaceous habitat. We provide an overview of the epidemiology, ecology and landscape characteristics of high risk environments for human acquisition of Lyme disease among 12 endemic counties in the U.S. State of Maryland. Reported Lyme disease cases were obtained from the Maryland Department of Health and Mental Hygiene for the 12 counties during 1996 – 2000. We geocoded patient residence. We used linear regression to model counts of reported Lyme disease by polygonal analysis units, which were delineated by federally funded roads. We applied a negative binomial distribution for our overdispersed count data, and applied a population offset derived from an overlay of US census 2000 data to assess rates of Lyme disease. Explanatory land use variables were derived from the 2001 National Landcover Database and intersected with analysis units; these included urban, water, forest and herbaceous landcovers. We used Fragstats and S Plus to analyze landcover and statistical data respectively. We report the model that provided the highest R2 value for explanatory variables with the lowest overall Akaike information criterion value. The most parsimonious explanatory model suggests that the highest risk landcover includes a 50:50 mix of forest and herbaceous cover with maximum common boundaries (edge) between the classes. The inclusion of income as a confounding variable improved the model diagnostic statistics. These findings are consistent with the known biological phenomenon of Lyme disease transmission, where vectors may thrive in degraded habitats. Results suggest that land use planning may provide a viable tool to reduce the risk for human acquisition of Lyme disease. The views expressed in this abstract are those of the individual authors and do not necessarily reflect the views and policies of the U.S. Environmental Protection Agency.