Traditional disease surveillance systems are instrumental in guiding policy-makers' decisions and understanding disease dynamics. The first study in this dissertation looks at sentinel surveillance network design. We consider three location-allocation models: two based on the maximal coverage model (MCM) and one based on the K-median model. The MCM selects sites that maximize the total number of people within a specified distance to the site. The K-median model minimizes the sum of the distances from each individual to the individual's nearest site. Using a ground truth dataset consisting of two million de-identified Medicaid billing records representing eight complete influenza seasons and an evaluation function based on the Huff spatial interaction model, we empirically compare networks against the existing volunteer-based Iowa Department of Public Health influenza-like illness network by simulating the spread of influenza across the state of Iowa. We compare networks on two metrics: outbreak intensity (i.e., disease burden) and outbreak timing (i.e., the start, peak, and end of the epidemic). We show that it is possible to design a network that achieves outbreak intensity performance identical to the status quo network using two fewer sites. We also show that if outbreak timing detection is of primary interest, it is actually possible to create a network that matches the existing network's performance using 42% fewer sites. Finally, in an effort to demonstrate the generic usefulness of these location-allocation models, we examine primary stroke center selection. We describe the ineffectiveness of the current self-initiated approach and argue for a more organized primary stroke center system. While these traditional disease surveillance systems are important, they have several downsides. First, due to a complex reporting hierarchy, there is generally a reporting lag; for example, most diseases in the United States experience a reporting lag of approximately 1-2 weeks. Second, many regions of the world lack trustworthy or reliable data. As a result, there has been a surge of research looking at using publicly available data on the internet for disease surveillance purposes. The second and third studies in this dissertation analyze Wikipedia's viability in this sphere. The first of these two studies looks at Wikipedia access logs. Hourly access logs dating back to December 2007 are available for anyone to download completely free of charge. These logs contain, among other things, the total number of accesses for every article in Wikipedia. Using a linear model and a simple article selection procedure, we show that it is possible to nowcast and, in some cases, forecast up to the 28 days tested in 8 of the 14 disease-location contexts considered. We also demonstrate that it may be possible in some cases to train a model in one context and use the same model to nowcast or forecast in another context with poor surveillance data. The second of the Wikipedia studies looked at disease-relevant data found in the article content. A number of disease outbreaks are meticulously tracked on Wikipedia. Case counts, death counts, and hospitalization counts are often provided in the article narrative. Using a dataset created from 14 Wikipedia articles, we trained a named-entity recognizer (NER) to recognize and tag these phrases. The NER achieved an F1 score of 0.753. In addition to these counts in the narrative, we tested the accuracy of tabular data using the 2014 West African Ebola virus disease epidemic. This article, like a number of other disease articles on Wikipedia, contains granular case counts and deaths counts per country affected by the disease. By computing the root-mean-square error between the Wikipedia time series and a ground truth time series, we show that the Wikipedia time series are both timely and accurate.

<p>Diss. (sammanfattning) Umeå : Umeå universitet, 2001. Härtill 6 uppsatser.</p> / digitalisering@umu

Thesis (M.A.)--University of South Florida, 2006. / Title from PDF of title page. Document formatted into pages; contains 86 pages. Includes bibliographical references.

Submitted by Luciana Ferreira (lucgeral@gmail.com) on 2018-04-16T13:26:11Z No. of bitstreams: 2 Dissertação - Jéssica Guimarães Rodrigues - 2018.pdf: 4231969 bytes, checksum: 7cd99fc018c0461eb571ee4a66eb3bbc (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2018-04-16T13:27:10Z (GMT) No. of bitstreams: 2 Dissertação - Jéssica Guimarães Rodrigues - 2018.pdf: 4231969 bytes, checksum: 7cd99fc018c0461eb571ee4a66eb3bbc (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2018-04-16T13:27:10Z (GMT). No. of bitstreams: 2 Dissertação - Jéssica Guimarães Rodrigues - 2018.pdf: 4231969 bytes, checksum: 7cd99fc018c0461eb571ee4a66eb3bbc (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2018-03-16 / Hemodialysis is the most common category of kidney replacement therapy set for chronical kidney disease. In order to perform this treatment it’s needed a vascular access (VA) that offers an adequate flow rate, a long use-life and a low rate of complications. The Arteriovenous Fistula (AVF) is the closest access to meet these requirements. It can, however, present complications and, during the cannulation that usually happens three times per week, adverse events (AE) can occur to the patient. The arteriovenous fistula cannulation must happen with safety in order to prevent future patency problems. There are three methods of cannulation: area, rope ladder and buttonhole. In the area method, the insertion points of the needles are in the same area; in the rope ladder method there’s the varying of the place of the puncture, at a distance defined by the previous puncture, all along the VA; and in the buttonhole method, the needle’s insertion happen in the same place, angle and deepness, forming a subcutaneous tunnel that will be cannulated with the blunt needle. Each one of these methods has its own particularity and can influence in the need to repair the fistula. This is a prospective cohort study, during the course of six months, from April to September of 2017, conducted with the participation of 347 patients using the vascular access by autologous arteriovenous fistula, within three hemodialysis clinics in the city of Goiânia - GO. The data collection happened by weekly interview to the patients, using a structured instrument online. The research was approved by the Ethics Committee and the participation conditioned to signing of the consent form by the patient. The general objective was to analyze the factors that can influence in the necessity to repair the arteriovenous fistula of patients in hemodialytic treatment cannulated by different cannulation methods. The specific objective was to relate the adverse events and complications in the different arteriovenous fistula cannulation methods. We’ve found that in the buttonhole method, the most frequent AE was dermatitis and misscannulation, and in the area/rope-ladder methods, the most frequent AE were haematoma and peri-punction bleeding. The patients in the buttonhole method group received the hemodialytic treatment with a higher blood flow compared the other group. We’ve observed that the dual lumen catheter (DLC) is a predictor to the need of AVF repairments, due to enhancing in 28% the risk of need for AFV repair. The “arterial” retrograde cannulation has presented itself as a protection factor, diminishing the need to AVF repairments in 1%. In conclusion, the buttonhole method is recommended, since there is an intermittent surveillance of the arteriovenous fistula by the nurse in the touching exam. The area method is not recommended, and the rope ladder method should be individually evaluated in future studies. The nurse must act by monitoring the AFV, surveillance of the patency parameters and health education to the patients for the AVF self-care, as well as continued education to the nursing team in order to promote safe and scientifically based practices. / A hemodiálise é a modalidade de terapia renal substitutiva mais comumente instituída para a doença renal crônica. Para esse tratamento é necessário um acesso vascular que ofereça fluxo sanguíneo adequado à necessidade dialítica, meia vida longa e baixo índice de complicações. A fístula arteriovenosa é o acesso que mais se aproxima desses requisitos. Porém, não obstante, pode apresentar complicações, e durante as punções, que comumente se repetem três vezes por semana, pode haver eventos adversos (EA) ao paciente. A punção da fístula arteriovenosa deve ser realizada com segurança a fim de prevenir futuros problemas de perviedade. Há três métodos de punção: regional, escada de corda, e buttonhole. No método regional, os pontos de inserção das agulhas são na mesma região; no método escada de corda, há rotação do sítio de punção, a uma distância definida a partir da anterior ao longo de todo o AV; e no buttonhole, a inserção da agulha é no mesmo local, ângulo e profundidade, formando de um túnel subcutâneo que será puncionado com agulha romba. Cada um desses métodos tem sua particularidade e podem influenciar na necessidade para reparos na fístula. Este é um estudo longitudinal de coorte prospectiva, no período de seis meses, abril a setembro de 2017, realizado com 347 pacientes em hemodiálise usando acesso vascular por fístula arteriovenosa autóloga, em três clínicas satélites do município de Goiânia - GO. A coleta de dados foi por entrevista semanal aos pacientes, por meio de instrumento estruturado online. A pesquisa foi aprovada por comitê de ética, e a participação condicionada à assinatura do Termo de Consentimento Livre e Esclarecido do paciente. O objetivo geral foi analisar fatores que influenciam na necessidade de reparo à fístula arteriovenosa de pacientes em hemodiálise puncionados por distintos métodos de punção. E os objetivos específicos foram identificar e relacionar os eventos adversos e complicações em distintos métodos de punção da fístula arteriovenosa, e caracterizar os preditores de complicações da fístula arteriovenosa. Encontramos como resultados que no método de punção de fístula arteriovenosa buttonhole o EA mais frequente foi dermatite e reinserção de agulhas de punção, e nos métodos escada/regional foram hematoma e sangramento peripunção. Os pacientes no grupo puncionado pelo método buttonhole receberam hemodiálise sob fluxos de sangue mais altos comparado ao outro grupo. Observamos que o uso do cateter venoso central de duplo lúmen (CDL) caracteriza-se um preditor de necessidade de reparo da fístula arteriovenosa, pois aumenta em 28% o risco dessa necessidade. A punção “arterial” retrógrada apresentou-se como fator de proteção, diminuindo em 1% a necessidade de reparos. Concluímos que o método de punção buttonhole é recomendado desde que haja a monitoração intermitente da fístula arteriovenosa pelo enfermeiro durante exame físico. O método regional é desestimulado. E o método escada de corda deve ser avaliado individualmente em estudos futuros. O enfermeiro deve estabelecer a vigilância dos parâmetros de perviedade, educação em saúde para autocuidado da fístula arteriovenosa, bem como educação continuada para a equipe de enfermagem a fim de promover práticas seguras e cientificamente embasadas.

Les maladies vectorielles sont en processus d’émergence à travers le monde. Au Canada, la maladie de Lyme (ML) a été identifiée comme une maladie infectieuse émergente prioritaire. La surveillance, si elle est efficace, peut suivre le portrait d’une maladie en évolution afin d'informer les autorités de santé publique; toutefois, en raison de ressources limitées, elle doit être optimisée. La surveillance sentinelle permet de réduire les coûts, car un nombre limité d'unités statistiques sont mesurées de manière répétée dans le temps. Néanmoins, ces unités sentinelles doivent être représentatives du portrait épidémiologique de la maladie pour assurer l'efficacité du système de surveillance. Le premier objectif de cette thèse était d'évaluer la représentativité de la surveillance sentinelle pour décrire le risque de ML au Québec, Canada. Deux types de systèmes de surveillance active acarologique sont déjà en place au Québec : 1) un système sentinelle, où les sites de terrain sentinelles sont maintenus constants et visités à chaque saison de terrain, et 2) un système de surveillance basé sur le risque où les sites accessoires sont priorisés en fonction de leur profil de risque. Des mesures de danger acarologique, en termes d'estimations de la densité de nymphes, ont été dérivées des sites sentinelles entre 2015 et 2019 et comparées à celles obtenues dans les sites accessoires. Les mesures de danger acarologique dérivées des sites sentinelles ont également été corrélées avec le nombre de cas humains rapportés à l'échelle municipale pour déterminer si elles étaient représentatives du risque de ML pour la population humaine. Il a été démontré que le système de surveillance sentinelle était capable de suivre les tendances spatio-temporelles d’incidence de ML dans les populations humaines de la zone d'étude et fournissait un meilleur indicateur de l’incidence de ML par rapport au système de surveillance basé sur le risque. Cependant, bien que les modèles aient pu prédire le risque de maladie de Lyme, les sites sentinelles n'ayant pas été choisis selon une approche validée, on peut émettre l’hypothèse que l’utilisation d'une approche holistique standardisée pour la sélection de sites sentinelles pourrait optimiser le design spatial du système de surveillance. Le deuxième objectif de cette thèse était de développer une telle approche et de l'appliquer à une étude de cas : la surveillance acarologique sentinelle pour le risque de ML à travers le Canada. Une revue de la portée a été utilisée pour inventorier les initiatives précédentes de surveillance sentinelle pour les maladies vectorielles, et pour cataloguer les critères qui ont été utilisés pour sélectionner les emplacements des unités sentinelles dans la zone d'étude. Les articles pertinents ont ensuite été analysés à l'aide d'une revue du type réaliste afin de créer un outil décisionnel permettant de sélectionner des critères pertinents pour la planification du design spatial d'un système de surveillance sentinelle pour les maladies vectorielles. Enfin, l'outil a été utilisé lors de la création d'un nouveau réseau de surveillance sentinelle pour le risque de ML au Canada; les critères retenus ont été incorporés dans une analyse multi-critères spatiale afin de sélectionner les régions sentinelles pour le réseau de surveillance acarologique active. Dans l'ensemble, cette thèse a exploré la surveillance sentinelle pour les maladies vectorielles, et a développé et testé une approche pour optimiser et standardiser la planification du design spatial des systèmes de surveillance sentinelle pour les maladies vectorielles. Dans de futurs travaux, cette approche devrait être mise en œuvre, évaluée et validée pour d’autres maladies et contextes épidémiologiques. / Vector-borne diseases are emerging all over the globe. In Canada, Lyme disease (LD) has been identified as a priority emerging infectious disease. Public health surveillance, if effective, can track disease risk to inform public health authorities; however, due to finite resources, it must be optimized. Sentinel surveillance can cut costs, since a limited number of statistical units are measured repeatedly through time. Nonetheless, these sentinel units must be representative of the risk landscape to ensure an effective surveillance system. The first objective of this thesis was to evaluate the representativeness of sentinel tick surveillance for the risk of LD in Québec, Canada. Two types of tick-based active surveillance systems are already in place in Québec: 1) a sentinel system, where sentinel field sites are kept constant and visited every field season, and 2) a risk-based surveillance system where accessory sites are prioritized according to their risk profile. Acarological hazard measures, in the form of nymph density estimates, were derived from sentinel sites between 2015 and 2019 and compared with those obtained from accessory sites. Hazard measures derived from sentinel sites were also correlated with LD incidence at the municipal scale to see if they were representative of LD risk to human populations. It was shown that the sentinel tick-based surveillance system was able to follow spatiotemporal LD incidence trends in human populations across the study zone and provided a better indicator of LD incidence in comparison with the risk-based surveillance system. However, as sentinel sites were not chosen using a validated approach, it can be hypothesized that the spatial design for the system could be optimized through the development of a standardized, holistic approach for sentinel site selection. The second objective of this thesis was therefore to develop such an approach and apply it to a case example: sentinel surveillance of LD across Canada. A scoping review was used to inventory previous sentinel surveillance initiatives for vector-borne diseases, and catalogue criteria which had been used to select sentinel unit locations across the study zone. Relevant papers were subsequently analyzed using a realist-type review to create a decision tool to select relevant criteria for planning the spatial design of a sentinel surveillance system for vector-borne diseases. Finally, the tool was applied to guide the creation of a new sentinel tick-based surveillance network for LD risk in Canada; the retained criteria were incorporated into a spatial multi-criteria decision analysis to select sentinel regions for the active surveillance network. Overall, this thesis has explored sentinel surveillance for vector-borne disease and has developed and applied an approach to optimize and standardize spatial design planning for vector-borne sentinel surveillance systems. In future work, this approach should be implemented, evaluated, and validated for other types of diseases and epidemiological contexts.

The literature directly comparing the epidemiology of MRSA among adult and pediatric hospitalized patients is strikingly minimal. The objective of this thesis was to identify any differences between these two patient groups. The Canadian Nosocomial Infections Surveillance Program MRSA data (1995 to 2007: n=1,262 pediatric and 35,907 adult cases) were used to compare MRSA clinical and molecular characteristics and rates. Hospital characteristics were modeled using repeated measures Poisson regressions. The molecular and epidemiological characteristics of MRSA differed significantly between adults and children. Compared to children, MRSA in adults was more likely to be healthcare-associated, colonization, SCCmec type II, PVL negative, and resistant to most antibiotics. Rates of MRSA in Canada increased in both populations over time but were significantly higher in adults. The hospital characteristics associated with increased MRSA rates differed in adult and pediatric facilities. Implications for infection prevention and control strategies are discussed.

The literature directly comparing the epidemiology of MRSA among adult and pediatric hospitalized patients is strikingly minimal. The objective of this thesis was to identify any differences between these two patient groups. The Canadian Nosocomial Infections Surveillance Program MRSA data (1995 to 2007: n=1,262 pediatric and 35,907 adult cases) were used to compare MRSA clinical and molecular characteristics and rates. Hospital characteristics were modeled using repeated measures Poisson regressions. The molecular and epidemiological characteristics of MRSA differed significantly between adults and children. Compared to children, MRSA in adults was more likely to be healthcare-associated, colonization, SCCmec type II, PVL negative, and resistant to most antibiotics. Rates of MRSA in Canada increased in both populations over time but were significantly higher in adults. The hospital characteristics associated with increased MRSA rates differed in adult and pediatric facilities. Implications for infection prevention and control strategies are discussed.