Perspectives in Lyme Disease Modeling

Srinath Krishnan

Lyme disease is one of the most frequently reported vector-borne diseases in the United States with an estimated 20,000 cases per year (Bacon et al., 2008). In the U.S., it was first identified in 1976 in a cluster of children in the town of Lyme, Connecticut (Steere, 1989). Initially diagnosed as juvenile rheumatoid arthritis, it was later reclassified as Lyme arthritis, a late manifestation of the illness now called Lyme disease. The disease is a zoonosis caused by the tick-borne bacterium Borrelia burgdorferi, a type of spirochete. Lyme is a multisystem inflammatory disease first affects the skin and, if untreated, can travel through the bloodstream and establish itself in joints, the nervous system and various body tissues. The identification of B. burgdorferi DNA in museum specimens of ticks and mice from Long Island suggests the prevalence of the disease during the late 19th and 20th century (Dennis and Hayes, 2002). B. burgdorferi, is maintained in a horizontal transmission cycle between its vector and the vertebrate reservoir hosts. The important vectors of the three pathogenic species of human Lyme borreliosis are the deer tick, Ixodes scapularis in the Northeast and central parts of the US; Western black-legged tick, Ixodes pacificus in the western US; the sheep tick, Ixodes rinius in Europe; and the taiga tick, Ixodes presulcatus in Asia.

I. scapularis has a two-year life cycle with three life stages; larvae, nymph and adult. It takes one blood meal per life stage and the two sub-adult stages are responsible for the enzootic maintenance of B. burgdorferi [Barbour and Fish, 1993]. Adult ticks feed and reproduce on white-tailed deer and other large and medium-sized mammals during autumn and early spring. They produce eggs in spring, which hatch in the summer. Larvae are most active during August and can acquire the disease pathogens if they feed on infected reservoir-competent hosts. Larvae and nymphs are host-generalists and feed on a variety of birds, mammals, and lizards (Lane et al., 1991).  Fed larvae drop off the host and molt into nymphs over winter. Nymphs emerge during spring and early summer and seek avian or mammalian hosts to feed on, including birds, small rodents, humans and pets. Due to the small size of the nymphs (the size of a poppy seed), they can go undetected in humans during the blood meal. Strong peaks in reported Lyme disease cases during the summer indicate that vector transmission from nymphs is responsible for the majority of Lyme disease cases. After feeding, nymphs transform into adults and emerge later in the year to quest for hosts.

Climate and habitat suitability models suggest that I. scapularis today occupy only a fraction of its suitable range. This range is predicted to expand under future warming (Ogden et al., 2006; Diuk-Wasser et al., 2010; Brownstein et al., 2003, 2005). Understanding how the distribution of the vector and associated pathogens will change in a warming world is crucial to prepare for public health risks with future climate change. The need for better predictive tools has led to numerous field monitoring campaigns and laboratory studies to examine the role of environmental and climatic variables. These relationships are used to drive geo-spatial, and mathematical models to predict tick, host, and human activity for a variety of future scenarios.

Spatial models are used to examine spatial changes in the factors that influence human incidence risk and the ecological components of Lyme disease risk; namely tick density and tick infection prevalence (see review in Killilea et al., 2008). The occurrence of Lyme disease in a region necessitates the close proximity of the pathogen, vector, one or more reservoir hosts and the human victim. But the factors that dictate the spatial distributions of each of these components are different. The distributions of ticks are determined by local temperature and humidity; distributions of small mammals determined by distribution of habitats providing abundant food, cover, and nesting sites; and human distributions are determined by socio-cultural factors and demographic trends. The probability of Lyme disease incidence will be highest in regions where conditions for vector, host, and human populations are favorable. For example, studies have shown the presence of forests in close proximity to residences is a good predictor of Lyme disease among members in the household (Glass et al., 1995; Kitron and Kazmierczak, 1997; Orloski et al., 1998; Eisen et al., 2006). This is supported by research that I. scapularis tick densities are generally higher in wooded areas and forests, compared to more open land types (Kitron et al., 1991; Maupin, 1991 Stafford and Magnarelli, 1993; Duffy et al., 1994). Similarly, it has also been observed that the prevalence of Lyme disease may be higher in edge environments where there is greater interaction between ticks and humans (Jackson et al., 2006). To model future changes in Lyme Disease, spatial models simulate landscape changes with future climate and apply observed relationships between land-type and Lyme disease to determine changes in the probability of vector distributions. While this approach can be used to generate high-resolution maps of future changes that are useful for developing planning strategies, it suffers from the underlying assumption that modern geospatial relationships between hosts, vectors, and human will remain similar in the future. Additionally, it should be noted that no study has clearly displayed the link between spatial variation in tick densities within or among specified habitat types. This suggests that climatic factors, rather than land-type are the primary determinant of tick activity. Diuk-Wasser et al. (2006) showed significant correlations between tick densities, temperature, relative humidity, and atmospheric pressure with tick density. Similar observations have led to the development of mathematical models to simulate various aspects of the complex tick lifecycle.

Mount et al. (1997) developed LYMESIM that simulated population dynamics of black-legged ticks and the transmission of associated vectors, to investigate the relationships between host density, tick density, and the incidence of the disease. Sandberg et al. (1992) used a matrix approach to study the seasonal interaction between ticks and various hosts. Ogden et al. (2006) simulated a population model and suggested that temperature was the primary factor that determined the suitability range of tick distributions in Canada. Changes in the suitability range with global warming were evaluated using climate information from models simulated to assess future changes in regional climate. Wu et al. (2013) modified the population model to enable the calculation of the basic reproductive number (R0), an important parameter used to evaluate the potential for the spread of the disease. One drawback in using these models for future projections is that temperature is the only climate variable used for estimating tick activity. This approach is further simplified by the use of the number of growing degree-days as a proxy for temperature and seasonality changes. While computationally less intensive, this simplification potentially limits the ability to simulate short-term changes in tick activity due to extreme weather events, predicted to increase in the future.

It is clear that the future of Lyme disease modeling lies in combining the salient features of these approaches. Direct coupling of the disease model with high-resolution regional climate and land-surface models will allow for the evaluation of spatial and temporal changes in tick distributions under different CO2 emission scenarios. Additionally, the coupling of host and human activity models to the disease model can provide a comprehensive picture of Lyme disease incidence on the regional scale. However, these results will be limited by the current lack of understanding of the relationships between climate drivers and tick activity rates. Therefore, substantial effort should be directed towards the collection of field and laboratory data to quantify the relationships between precipitation, humidity and tick activity at various stages in the tick lifecycle. The effect of uncertainties in climate modeling estimates on disease predictions should also be thoroughly tested using a variety of global and regional scale climate models. Ultimately, the goal of these efforts should be to obtain a range of best and worst-case scenarios, so that suitable disease prevention and adaptation strategies can be established across different spatial scales.

References

Bacon, R.M., Kugeler, K.J., and Mead, P.S. 2008. Surveillance for Lyme Disease - United States, 1992-2006. Center for Disease Control and Prevention. http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5710a1.htm.

Barbour, A. G., Fish, D. 1993. The Biological and Social Phenomenon of Lyme-Disease. Science, 260, 1610-1616.

Brownstein, J.S., Holford, T.R., Fish, D., 2003. A climate-based model predicts the spatial distribution of the Lyme disease vector Ixodes scapularis in the United States. Environmental Health Perspectives, 111, 1152–1157.

Brownstein, J.S., Skelly, D.K., Holford, T.R., Fish, D., 2005. Forest fragmentation predicts local scale heterogeneity of Lyme disease risk. Oecologia, 146, 469–475.

Dennis, D.T., Hayes, E.B., 2002. Epidemiology of Lyme Borreliosis. In Lyme borreliosis: biology, epidemiology and control. CABI Publishing. Oxford, United Kingdom. 251–280. 


Diuk-Wasser, M.A., Gatewood, A., Cortiñas, R., Yaremych-Hamer, S., Tsao, J., Kitron, U., Hickling, G., Brownstein, J., Walker, E., Piesman, J., and Fish, D., 2006. Spatiotemporal patterns of nymphal host-seeking I. scapularis (Acari: Ixodidae) in the United States. Journal of Medical Entomology, 43, 166-176.

Diuk-Wasser, M.A., Vourc’h, G., Cislo, P., Hoen, A.G., Melton, F., Rowland, M., Cortinas, R., Hickling, G.J., Tsao, J.I., Barbour, A.G., Kitron, U., Piesman, J., and Fish, D., 2010. Field and climate based model for predicting the density of host-seeking nymphal Ixodes scapularis, an important vector of tick-borne disease agents in the eastern United States. Global Ecol Biogeogr 19:504-514.

Duffy, D.C., Clark, D.D., Campbell, S.R., Gurney, S., Perello, R., Simon, N., 1994. Landscape patterns of abundance of Ixodes-scapularis (Acari, Ixodidae) on Shelter Island, New York. Journal of Medical Entomology 31, 875–879.

Eisen, R.J., Eisen, L., Lane, R.S., 2006. Predicting density of Ixodes pacificus nymphs in dense woodlands in Mendocino County, California, based on geographic information systems and re- mote sensing versus field-derived data. American Journal of Tropical Medicine and Hygiene, 74, 632–640.

Glass, G.E., Schwartz, B.S., Morgan, J.M., Johnson, D.T., Noy, P.M., Israel, E., 1995. Environmental risk factors for Lyme-disease identified with geographic information-systems. American Journal of Public Health, 85, 944–948.

Jackson, L.E., Levine, J.F., Hilborn, E.D., 2006. A comparison of analysis units for associating Lyme disease with forest edge habitat. Community Ecology, 7, 189–197.

Killilea, M.E., Swei, A., Lane, R.S., Briggs, C.J., Ostfeld, R.S., 2008. Spatial dynamics of Lyme disease: A review. Ecohealth 5: 167-195.

Kitron, U., Bouseman, J.K., Jones, C.J., 1991. Use of the Arc Info GIS to study the distribution of Lyme-disease ticks in an Illinois county. Preventive Veterinary Medicine 11, 243–248.

Kitron, U., Kazmierczak, J.J., 1997. Spatial analysis of the distribution of Lyme disease on Wisconsin. American Journal of Epidemiology, 145, 558–566.

Lane R.S., Brown, R.N., 1991. Wood rats and kangaroo rats: potential reservoirs of the Lyme disease spirochete in California. Journal of Medical Entomology, 28, 299–302.

Maupin, G.O., Fish, D., Zultowsky, J., Campos, E.G., Piesman, J., 1991. Landscape ecology of Lyme disease in a residential area of Westchester County, New York. American Journal of Epidemiology, 133, 1105–1113.

Mount, G.A., Haile, D.G., Daniels, E., 1997. Simulation of blacklegged tick (Acari: Ixodidae) population dynamics and transmission of Borrelia burgdorferi. Journal of medical entomology, 34.4, 461-484.

Ogden, N.H., Maarouf, A., Barker, I.K., Bigras-Poulin, M., Lindsay, L.R., Morshed, M.G., O’Callaghan, C.J., Ramay, F., Waltner-Toews, D., and Charron, D.F,  2006,  Climate change and the potential for range expansion of the Lyme disease vector Ixodes scapularis in CanadaInternational Journal of Parasitology 3, 63–70.

Orloski, K.A., Campbell, G.L., Genese, C.A., Beckley, J.W., Schriefer, M.E., Spitalny, K.C., Dennis, D.T., 1998. Emergence of Lyme disease in Hun- terdon County, New Jersey, 1993: a case-control study of risk factors and evaluation of reporting patterns. American Journal of Epidemiology, 147, 391–397.

Sandberg, S., Awerbuch, T.E., Spielman, A., 1992. A comprehensive multiple matrix model representing the life cycle of the tick that transmits the age of Lyme disease. J. Theor. Biol. 157, 203–220.

Stafford, K.C., Magnarelli, L.A., 1993. Spatial and temporal patterns of Ixodes-scapularis (Acari, Ixodidae) in southeastern Connecticut. Journal of Medical Entomology, 30, 762–771.

Steere, A.C., 1989. Lyme disease. N. Engl. J. Med. 321, 586–596.

Wu, X., Duvvuri, V.R., Lou, Y., Ogden, N.H., Pelcat, Y., Wu, J., 2013. Developing a temperature-driven map of the basic reproductive number of the emerging tick vector of Lyme disease Ixodes scapularis in Canada. Journal of theoretical biology, 319, 50-61.