A community modeling approach to understanding the ecology of Lyme disease

Halsey, Samniqueka Joi-Weaver

Permalink

https://hdl.handle.net/2142/104968 Copy

Description

Title

A community modeling approach to understanding the ecology of Lyme disease

Author(s)

Halsey, Samniqueka Joi-Weaver

Issue Date

2019-02-21

Director of Research (if dissertation) or Advisor (if thesis)

Miller, James R.

Doctoral Committee Chair(s)

Miller, James R.

Committee Member(s)

• Allan, Brian F.
• O'Dwyer, James
• Westervelt, James

Department of Study

School of Integrative Biology

Discipline

Ecol, Evol, Conservation Biol

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Disease
• Modelling
• Lyme disease

Abstract

The black-legged tick (Ixodes scapularis) is responsible for transmitting the Lyme disease pathogen (Borrelia burgdorferi) to humans in eastern North America. Since the 1970s, the geographic distributions of both the pathogen and the black-legged tick have expanded throughout the US. This expansion is thought to be primarily due to highly mobile wildlife hosts such as migratory birds and white-tailed deer (Odocoileus virginianus), although the relative contribution of each host has yet to be quantified. Previous approaches to understanding the ecology of Lyme disease, the most commonly reported tick-borne disease, have been limited to determining the roles that individual wildlife hosts play in the dispersal, transmission, and maintenance of B. burgdorferi and I. scapularis populations. In my research, I use individual-based models to depict the complex interactions of this multi-host disease system. Understanding and controlling this tick-borne disease in humans requires going beyond individual roles and embracing the complex interactions between the tick, pathogen, and wildlife hosts. Due to the ongoing expansion of I. scapularis, there is a need to identify the role wildlife hosts play in the establishment and maintenance of tick populations. To quantify and synthesize the patterns of I. scapularis and B. burgdorferi prevalence relative to wildlife hosts, I reviewed the findings of independent studies conducted throughout the United States. I performed a comprehensive literature review covering the period 1970-2017 as part of a meta-analysis of individual wildlife hosts captured and examined for I. scapularis and subsequently tested for B. burgdorferi. I investigated whether there were regional differences in tick infestation and pathogen prevalence between the Northeast, Midwest and the Southeast U.S. using generalized linear models. In most cases, detection of I. scapularis and B. burgdorferi was significantly higher in the Northeast than the Midwest. Using these data, I developed an epizootiological model to determine the relative contributions of individual hosts to B. burgdorferi-infected nymphs, providing additional evidence that wildlife hosts other than white-footed mice (Peromyscus leucopus) may contribute to Lyme disease risk. This research identified key directions for future research, including a greater focus on non-mice species and on wildlife hosts that have been poorly sampled in the past. Using data reported in the scientific literature, I designed a spatially explicit individual-based tick interaction model (SEIB-TIM) that uses a bottom-up approach to examine the processes through which I. scapularis populations are maintained. The maintenance of tick populations can be better understood, and controlling mechanisms identified, when all elements of the tick life-cycle are incorporated in the model. Using a two-host wildlife community consisting of mice and deer, I parameterized the model so that tick infestation rates for mice are within the range of those reported in field studies. Once the SEIB-TIM accurately simulated the interactions between I. scapularis, wildlife hosts, and the environment, I evaluated its robustness to parameter uncertainty using both global and local sensitivity analyses. Lastly, I related changes in model parameters to tick life-history traits to understand how those changes affect the maintenance of I. scapularis populations for ten years. Results from this modeling exercise indicate that interventions aimed at both decreasing the number of larval ticks that can successfully feed on mice. Further, management efforts will be most effective in reducing tick populations when targeting the reproductive stage of the ticks’ life cycle on deer. This said, it could take five to ten years for a reduction in tick populations to manifest. I therefore conclude that management should target multiple stages in the tick’s life cycle and over the long term. Lastly, the SEIB-TIM serves as the foundation for a more complex model that includes four additional host species and the Lyme disease pathogen B. burgdorferi. The emergence of Lyme disease in the Northeast is attributed to reforestation of the regions and subsequent lack of species diversity among wildlife hosts. Increased biodiversity can regulate the abundance and distribution of a pathogen, a phenomenon known as a “dilution effect”. The premise of the dilution effect is that additional species diminishes the probability of pathogen transmission to the vector by the most competent disease reservoir. Several mechanisms (e.g., vector regulation, encounter reduction, transmission reduction) have been proposed to explain the relationship between biodiversity and disease. Due to the infeasibility of conducting large-scale experiments that manipulate wildlife host community composition, modeling can enhance our understanding of the mechanisms driving patterns seen in nature. The SEIB-TIM allows me to vary the presence and densities of wildlife hosts to explicitly test proposed mechanisms driving changes in pathogen prevalence as a consequence of host community composition, including species richness, evenness, and host density. My results suggest that while increasing both species richness and Shannon H diversity corresponds to an increased proportion of ticks infected by the pathogen, the overall density of infected nymphs decreases, supporting both vector regulation and transmission reduction. In addition, increasing species richness reduces the number of nymphs fed by the mice, the most competent disease reservoir, providing support for the encounter reduction mechanism. These results have several implications for managing Lyme disease risk including being able to predict how changes in host community composition influence pathogen transmission cycles in nature. My overall research objective is to advance our understanding of the mechanisms involved in the invasion and establishment of I. scapularis and B. burgdorferi by utilizing a community framework via individual-based modeling. By answering the call for a more integrated, community-based approach, this modeling approach enables me to identify the mechanisms underlying the spatial and temporal patterns of tick abundance and pathogen prevalence in the Lyme disease system. This modeling technique may be broadly applicable to other emerging diseases, facilitating more informed decisions in the development of wildlife management plans, and interventions to predict and mitigate disease risk.

Graduation Semester

2019-05

Type of Resource

text

Permalink

http://hdl.handle.net/2142/104968

Copyright and License Information

Copyright 2019 Samniqueka Halsey

Owning Collections