Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
6 results
Search Results
Item Associations between Urinary Phthalates and Metabolic Syndrome in NHANES 2005-2010(2015) Haque, Mefruz Salwa; Dallal, Cher M; Epidemiology and Biostatistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Phthalates, commonly used to make plastics more durable, are a group of endocrine disrupting chemicals (EDC), with potential for adverse metabolic consequences. Associations between exposure to 13 phthalate metabolites and the prevalence of metabolic syndrome (MetS) were examined among 5,409 U.S adults ≥ 18 years of age, who participated in the National Health and Nutrition Examination Survey from 2005-2010. MetS was assessed using clinical and questionnaire data. Odds Ratio (OR) and 95% Confidence Intervals (CI) adjusting for age, creatinine and key confounders, were estimated with multivariable logistic regression. Positive associations were observed between individual phthalate metabolites and MetS: (MCOP OR=1.31, 95% CI=1.40, 1.64, p-trend<.01; MCPP OR=1.39, 95% CI=1.09, 1.77, p-trend=0.01). In gender stratified analyses, findings with MCOPP and MCPP were restricted to women only. Phthalate metabolites may increase the prevalence of MetS; however, further studies are needed to better understand the role of EDCs in the development of MetS.Item International Externalities in Pandemic Influenza Mitigation(2011) Hutton, Stephen; Cropper, Maureen; Economics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)A serious influenza pandemic could be devastating for the world. Ideally, such a pandemic could be contained, but this may be infeasible. One promising method for pandemic mitigation is to treat infectious individuals with antiviral pharmaceuticals. While most of the benefits from treatment accrue to the country in which treatment occurs, there are some positive spillovers: when one country treats more of its population this both reduces the attack rate in the other country and increases the marginal benefit from additional treatment in the other country. These externalities and complementarities may mean that self-interested rich countries should optimally pay for some AV treatment in poor countries. This dissertation demonstrates the presence of antiviral treatment externalities in simple epidemiological SIR models, and then in a descriptively realistic Global Epidemiological Model (GEM). This GEM simulates pandemic spread between cities through the international airline network, and between cities and rural areas through ground transport. Under the base case assumptions of moderate transmissibility of the flu, the distribution of antiviral stockpiles from rich countries to poor and lower middle income countries may indeed pay for itself: providing a stockpile equal to 1% of the population of poor countries will reduce cases in rich countries after 1 year by about 6.13 million cases at a cost of 4.62 doses per rich-country case avoided. Concentrating doses on the outbreak country is, however, even more cost-effective: in the base case it reduces the number of influenza cases by 4.76 million cases, at the cost of roughly 1.92 doses per case avoided. These results depend on the transmissibility of the flu strain, the efficacy of antivirals in reducing infection and on the proportion of infectious who can realistically be identified and treated.Item USING AND MANIPULATING PROBABILISTIC CONNECTIVITY IN SOCIAL NETWORKS(2011) DuBois, Thomas M.; Srinivasan, Aravind; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Probabilistic connectivity problems arise naturally in many social networks. In particular the spread of an epidemic across a population and social trust inference motivate much of our work. We examine problems where some property, such as an infection or influence, starts from some initially seeded set of nodes and every affected node transmits the property to its neighbors with a probability determined by the connecting edge. Many problems in this area involve connectivity in a random- graph - the probability of a node being affected is the probability that there is a path to it in the random-graph from one of the seed nodes. We may wish to aid, disrupt, or simply monitor this connectivity. In our core applications, public health officials wish to minimize an epidemic's spread over a population, and connectivity in a social network suggests how closely tied its users are. In support of these and other applications, we study several combinatorial optimization problems on random-graphs. We derive algorithms and demonstrate their effectiveness through simulation, mathematical proof, or both.Item Anomaly Detection in Time Series: Theoretical and Practical Improvements for Disease Outbreak Detection(2009) Lotze, Thomas Harvey; Shmueli, Galit; Applied Mathematics and Scientific Computation; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The automatic collection and increasing availability of health data provides a new opportunity for techniques to monitor this information. By monitoring pre-diagnostic data sources, such as over-the-counter cough medicine sales or emergency room chief complaints of cough, there exists the potential to detect disease outbreaks earlier than traditional laboratory disease confirmation results. This research is particularly important for a modern, highly-connected society, where the onset of disease outbreak can be swift and deadly, whether caused by a naturally occurring global pandemic such as swine flu or a targeted act of bioterrorism. In this dissertation, we first describe the problem and current state of research in disease outbreak detection, then provide four main additions to the field. First, we formalize a framework for analyzing health series data and detecting anomalies: using forecasting methods to predict the next day's value, subtracting the forecast to create residuals, and finally using detection algorithms on the residuals. The formalized framework indicates the link between the forecast accuracy of the forecast method and the performance of the detector, and can be used to quantify and analyze the performance of a variety of heuristic methods. Second, we describe improvements for the forecasting of health data series. The application of weather as a predictor, cross-series covariates, and ensemble forecasting each provide improvements to forecasting health data. Third, we describe improvements for detection. This includes the use of multivariate statistics for anomaly detection and additional day-of-week preprocessing to aid detection. Most significantly, we also provide a new method, based on the CuScore, for optimizing detection when the impact of the disease outbreak is known. This method can provide an optimal detector for rapid detection, or for probability of detection within a certain timeframe. Finally, we describe a method for improved comparison of detection methods. We provide tools to evaluate how well a simulated data set captures the characteristics of the authentic series and time-lag heatmaps, a new way of visualizing daily detection rates or displaying the comparison between two methods in a more informative way.Item Mycobacteriosis in Chesapeake Bay striped bass Morone saxatilis(2008-04-22) Stine, Cynthia Bee; Kane, Andrew S.; Marine-Estuarine-Environmental Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Striped bass, Morone saxatilis, is an economically and ecologically important species in the Chesapeake Bay and along the East coast of the United States. In 1997 an epizootic of mycobacterial infections was discovered in the Chesapeake Bay stock and subsequent reports indicated that up to three-fourths of subpopulations of striped bass in the Bay were infected, primarily older fish. This study investigated regional and age class differences in mycobacterial infections among younger striped bass in the Chesapeake Bay, and identified putative risk factors for infection. Approximately 2,000 0+ to 3+ age striped bass, a limited number of spawning stock, and bycatch species were evaluated for microbiology, histopathology and parasitology. Mycobacterial isolates were grouped according to gas chromatography fatty-acid methyl-ester profiles and multi-locus sequencing. Twenty-nine groups of mycobacteria were discerned including M. scrofulaceum, M. septicum, M. interjectum, M. triplex/M. montefiorense, M. szulgai, M. moriokaense, M. duvalii, M. avium, M. terrae, M. pseudoshottsii/M. marinum and M. shottsii, and several putative new species. The majority of mycobacteria groups observed had host overlap. Data revealed that prevalence of mycobacterial infection increased with age, up to 59%. Location of capture was associated with higher infection prevalence in fish sampled from the Pocomoke River compared with fish sampled from the Upper Bay (1+), the Choptank River (1+) and the Potomac River (0+, 1+). The presence of copepods, isopods, acanthacephalans, nematodes and trichodinid ciliates was associated with an increased prevalence odds ratio (POR) for mycobacterial infection, while the presence of bacteria other than mycobacteria was associated with a decreased POR for 0+ fish. Gender was not a risk factor for mycobacterial infection, however, gonads from some mature fish were infected. In addition, mycobacterial infections were observed in 12 other Chesapeake Bay fishes, including Atlantic menhaden, Brevoortia tyrannus, an important prey species. Mycobacterial infections in Chesapeake Bay fish appear to be more complex than the one pathogen-one host scenario. Further, vertical and food-borne transmission cannot be ruled out. Future research requires an holistic approach including evaluation of multiple host species in association with water quality and other environmental parameters.Item EPIDEMIOLOGIC ANALYSIS OF RISK FACTORS FOR LOCAL DISAPPEARANCES OF NATIVE RANID FROGS IN ARIZONA(2005-08-11) Witte, Carmel Lee; Kane, Andrew S.; Animal Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This study used epidemiologic case-control methodology to examine habitat and environmental factors contributing to amphibian declines in Arizona. Risk factors were compared between sites where frogs disappeared (cases) and persisted (controls) using univariate and multivariable logistic regression analyses. Thirty-six percent (117/324) of all sites became cases during the study period. Elevation, non-native predators, hydrologic characteristics, aspect, and effects of nearby sites were significantly associated with frog persistence or disappearance. In the final multivariable model, risk for disappearance increased with increasing elevation (OR=2.7 for every 500 meters, P<0.01). Sites where disappearances occurred were 4.3 times more likely to have other nearby sites that also experienced disappearances (P<0.01), while having an extant population nearby decreased risk of disappearance by 85% (OR=0.15, P<0.01). Sites experiencing disappearances were 2.6 times more likely to have crayfish than control sites (P=0.04). Identification of risk factors associated with frog disappearances will guide future research and conservation efforts.