Files in this item

FilesDescriptionFormat

application/pdf

application/pdfKARKI-DISSERTATION-2017.pdf (3MB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:Improved estimates of risk of exposure to pathogens transmitted by mosquitoes
Author(s):Karki, Surendra
Director of Research:O'Hara Ruiz, Marilyn
Doctoral Committee Chair(s):O'Hara Ruiz, Marilyn
Doctoral Committee Member(s):McLafferty, Sara; Li, Bo; Smith, Rebecca Lee; Harbison, Justin; Westcott, Nancy
Department / Program:Pathobiology
Discipline:VMS - Pathobiology
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Catch basins
Climate
Chicago
Culex mosquitoes
Demographic characteristics
Illinois
Mosquito abundance
Mosquito infection
Mosquito infection rate
Mosquito-borne diseases
Multi-level modeling
Vector-borne diseases
West Nile virus, Spatial epidemiology
Trapping methods
Weather
Landscape
Abstract:The exposure of humans to mosquitoes that carry infectious pathogens is a central component to assess the risk of people becoming infected with the pathogen and succumbing to a range of illness. Mosquito-borne illnesses are increasing, but the understanding of the effect of exposure to mosquitoes on the variability of risk has not always been clear. The risk of exposure to mosquito-borne pathogens is complicated by the transmission cycle of the pathogen, the dynamic nature of conditions that affect mosquito abundance and species- and regionally-specific mosquito biting behavior. In the United States, West Nile virus (WNV), transmitted by several species of Culex mosquitoes, has been the leading cause of mosquito-borne illness since its first introduction in 1999. The state of Illinois first encountered West Nile virus in 2001, and the first human cases were reported in 2002. The state has seen significant spatial and temporal variation in WNV cases since then. The overall objective of this dissertation work was to improve our understanding of spatial variability of risk for transmission of WNV to humans. We exploited long-term data on mosquito collection and testing, human WNV illness, weather, landscape, and demographic factors, and used statistical and geospatial approaches to address questions related to the factors that drive vector mosquito abundance, mosquito infection, and human WNV illness. We evaluated the local weather and landscape factors associated with Culex abundance first independently, and later used multilevel modeling approach to evaluate the joint effects and weather and landscape when both are analyzed together. We hypothesized that the estimates of mosquito abundance are affected by the trapping methods used to capture them, and we considered the degree to which this factor needs to be taken in to account when analyzing mosquito abundance data. Further, we developed weather based weekly prediction models for the WNV mosquito infection rate (MIR) for the state of Illinois and nine climate divisions. We observed that the MIR model performed better for northeastern Illinois where intensive mosquito surveillance is carried out compared to southern parts of the Illinois. Finally, we determined the fine-scale dynamic drivers of spatiotemporal variability in human WNV cases in the Chicago region. Using mixed-effects multiple logistic regression analysis, we identified that hot and dry weather conditions and higher mosquito infection rate in preceding weeks were the main drivers of spatiotemporal variability of human WNV illness in Chicago area, with some demographic and landscape characteristics contributing to it. In conclusion, our study helped to understand several essential factors associated with vector mosquito abundance, WNV mosquito infection, and human WNV illness, thus improving our understanding of the risk for pathogens transmitted by mosquitoes at fine geographical and temporal scales. We also demonstrated that the long-term surveillance data on mosquito data and human illness data coupled with publicly available weather, landscape, and demographic data can be successfully used to understand the drivers of disease and to develop prediction models. The knowledge we gained from our approach can be extrapolated to understand the spatial epidemiology of other mosquito-borne diseases, such as St. Louis encephalitis, dengue and chikungunya.
Issue Date:2017-12-05
Type:Thesis
URI:http://hdl.handle.net/2142/99233
Rights Information:Copyright 2017 Surendra Karki
Date Available in IDEALS:2018-03-13
Date Deposited:2017-12


This item appears in the following Collection(s)

Item Statistics