The incidence and geographic range of Lyme disease continues to increase in the United States because of the expansion of Ixodes scapularis, the main Lyme disease vector, as well as other emerging tick species and tick-borne pathogens. This talk is part of a symposium exploring the use tick and tick pathogen surveillance to track the expanding risk of tick-borne diseases. I review different surveillance data types and how they have been used to generate maps of tick/pathogen habitat suitability and predict the distribution of human infection risk and disease cases. I introduce a dynamic modeling approach used to predict the probability of reporting a first case of Lyme disease using static tick occurrence estimates and environmental predictors. The model predicted a county’s first reported Lyme disease case a mean (SD) of 5.5 (3.5) years earlier than was reported to the CDC, with a mean spread velocity estimated at 27.4 (95%CI, 13.6-54.4) km per year. The estimated mean time lag between the first reported case in a neighboring county and any county was 7 (95%CI, 3-8) years. Using a similar modeling approach, we explored the use of time series of passive and active tick surveillance to predict Lyme disease incidence at the census tract level in New York State. Both passive and active tick surveillance provided similar predictive power, although passive surveillance provided a more complete coverage and may thus be a more effective approach.
