Browsing by Subject "kernel density estimation"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A comprehensive look at malaria in an epidemic prone area of highland Kenya using a traditional, spatial, and immuno-epidemiologic approach
    (2015-10) Hamre, Karen
    Malaria is the leading cause of morbidity and mortality in Kenya. In highland areas with unstable transmission, malaria occurs among individuals of all ages as immunities are not built or sustained with limited exposure to bites from infected mosquitoes. With little immunity, these populations are susceptible to epidemics; about 20% of an estimated 45.9M Kenyans live in highland settings. The Kenya Medical Research Institute-University of Minnesota Malaria Research Project has collected prospective data on a cohort in highland Kipsamoite and Kapsisiywa since 2003, and was the source of data and samples used for this dissertation. A 13-month period of possible interruption of clinical malaria was reported in this area after the Ministry of Health (MOH) implemented indoor residual spraying campaigns and switched to first-line drugs for treatment. HASH(0x7febe3cbef18) Initially, the MOH National Malaria Strategy was evaluated where monthly incidence and intervention coverage levels were determined from the period after interruption through 2013. Then, a focused analysis of the MOH-led mass insecticide-treated bed net (ITN) distribution campaign was done where generalized estimating equations were used to evaluate the impact of the campaign itself on malaria incidence, and multi-level mixed effects logistic regression was used to evaluate whether individual bed net use was associated with incidence. Both analyses showed some benefit that approached, but did not reach statistical significance. However, universal coverage of ITN, while targeted in the mass campaign, was not achieved in the study area so it is unknown whether truly complete coverage of ITN in the area would have led to a greater decrease in transmission. HASH(0x7febe3ca4e30) Next, kernel density estimation methods for analyzing spatial point patterns were used to evaluate the spatial variation of clinical malaria during periods from 2003–2013, before and after interruption, and for periods of peak incidence. Heat maps were created to illustrate the estimated probabilities and relative risks, and were overlaid with asymptotic tolerance contours to identify areas with significantly elevated risk of malaria. The regions of elevated risk were not consistent during peak periods before and after interruption of clinical malaria cases, likely due to asymptomatic transmission still circulating in the study area. Eliminating malaria in this setting will likely require a multi-stage approach. Finally, a nested matched case-control study was designed to assess correlates of protection from clinical malaria in a low transmission setting, which is inherently difficult. Conditional logistic regression was used to evaluate the associations of antibody responses to 2 pre-erythrocytic and 9 blood stage antigens with the odds of developing clinical malaria. Four possible candidate antigens were identified that could inform novel vaccine candidate combinations. Evaluating the epidemiology of malaria in this setting is especially important with the renewed interest among the international community to eradicate malaria, as intervention measures will increase and stable transmission patterns will likely convert to unstable, leaving large new populations living in transmission intensities and having similar immunologic profiles as our highland Kenya cohort.
  • Thumbnail Image
    Item
    Data, R Code, and Output Supporting: A “How-to” Guide for Estimating Animal Diel Activity Using Hierarchical Models
    (2024-10-22) Iannarilli, Fabiola; Gerber, Brian D; Erb, John; Fieberg, John R; jfieberg@umn.edu; Fieberg, John; Fieberg Lab
    This repository contains data, code, and associated output supporting work presented in "Iannarilli F., Gerber B. D., Erb J., and Fieberg J. R. (2024). A “How-to” Guide for Estimating Animal Diel Activity Using Hierarchical Models. Journal of Animal Ecology". We provide a series of .Rmd files that can be compiled to form a step-by-step tutorial demonstrating how to quantify animal activity patterns from time-stamped data using trigonometric and cyclic cubic spline hierarchical models. These models can accommodate site-to-site variability in the frequency of site use and timing of activity, while accounting for sampling effort. The text is accompanied by a series of examples in which we address common ecological questions related to the study of animal diel activity, such as the shape of the underlying activity pattern (unimodal, bimodal, or cathemeral) and the effect of covariates or the co-occurrence of other species on activity patterns. An HTML version of this tutorial is available at https://hms-activity.netlify.app/.