Submitted by admin tede (tede@pucgoias.edu.br) on 2018-11-19T16:59:56Z No. of bitstreams: 1 GISELLE ANGÉLICA MOREIRA DE SIQUEIRA.pdf: 1650143 bytes, checksum: 0a85747b640c1ee6d3c2dc446dececf7 (MD5) / Made available in DSpace on 2018-11-19T16:59:56Z (GMT). No. of bitstreams: 1 GISELLE ANGÉLICA MOREIRA DE SIQUEIRA.pdf: 1650143 bytes, checksum: 0a85747b640c1ee6d3c2dc446dececf7 (MD5) Previous issue date: 2013-12-19 / SIQUEIRA, Giselle Angelica Moreira de. Epidemic Influenza A (H1N1) 2009 in the state of Goiás/Brazil: cases and deaths. Dissertation (MSc in Environmental Sciences) – Catholic University of Goiás, Goiânia, 2013. Between late March and early April 2009, were the first reported cases of human infection caused by a new viral subtype Influenza A (H1N1) in Southern California and near San Antonio, Texas, USA, and then in Mexico and Canada. Until July 6, 2009, 905 cases were confirmed by the Ministry of Health, with reports of 23 states and the Federal District. This study described the profile of confirmed cases and deaths affected by Influenza A ( H1N1 ) in 2009 in the state of Goias and Brazil through a descriptive ecological study of confirmed cases and deaths affected by Influenza A virus (H1N1) 2009 in the State of Goias and Brazil between epidemiological weeks 16 th to 52 th, variables of research Influenza record, feeding SINAN Influenza Web were selected such as epidemiological week, age, gender, education, signs and symptoms, comorbidities, vaccination status, hospitalizations and evolution. Among the total number of cases reported during the epidemic , more than 45% were confirmed Influenza A (H1N1) in Goiás and in Brazil , with 14.9% and 3.9% subsequently died respectively. Females were predominant, those over 6 % were pregnant. The age range was found between 15 and 45 years, with the primary and secondary school levels observed schooling. Among the signs and symptoms , more than 95% of cases and deaths had fever, cough and dyspnoea, less than 30% had comorbid conditions, the occurrence of hospitalizations of cases was 96% and 45% in Goiás in Brazil, while hospitalization those who subsequently died was above 96%, less than 14% of cases and deaths have taken the vaccine against influenza (H1N1). It was concluded that it was possible to know the profile of cases and deaths from socio demographic and clinical characteristics during the epidemic period Influenza (H1N1) 2009 in Goias and Brazil, many lessons were learned that will assist in the consolidation of plans to tackle the unusual situations of epidemic and pandemic character and guide the development of public policies that will strengthen the surveillance system of disease, health care, implementation of laboratory diagnosis, mass vaccination and personal protection and respiratory hygiene network. / SIQUEIRA, Giselle Angélica Moreira de. Epidemia da Influenza A (H1N1) 2009 no estado de Goiás/Brasil: casos e óbitos. Dissertação (Mestrado em Ciências Ambientais) – Pontifícia Universidade Católica de Goiás, Goiânia, 2013. Entre o final de março e começo de abril de 2009, foram notificados os primeiros casos de infecção humana causada por um novo subtipo viral Influenza A (H1N1), no sul da Califórnia e próximo de San Antonio, no Texas, Estados Unidos, e, em seguida, no México e Canadá. Até o dia 06 de julho de 2009, 905 casos foram confirmados pelo Ministério da Saúde, com notificações de 23 estados e do Distrito Federal. Neste estudo foi descrito o perfil dos casos confirmados e óbitos acometidos por Influenza A (H1N1) em 2009 no Estado de Goiás e Brasil por meio de um estudo ecológico descritivo dos casos confirmados e óbitos acometidos pelo vírus Influenza A (H1N1) 2009 no Estado de Goiás e Brasil entre as semanas epidemiológicas 16ª a 52ª, foram selecionadas variáveis da ficha de investigação de Influenza, que alimenta o SINAN Influenza Web tais como semana epidemiológica, faixa etária, gênero, escolaridade, sinais e sintomas, comorbidades, situação vacinal, hospitalizações e evolução. Dentre o total de casos notificados durante a epidemia, mais de 45% foram confirmados por Influenza A (H1N1) em Goiás e no Brasil, sendo que 14,9% e 3,9% evoluíram para o óbito respectivamente. O gênero feminino foi predominante, destas mais de 6% eram gestantes. A faixa etária encontrada foi entre 15 a 45 anos, sendo o ensino médio e fundamental os níveis de escolaridade constatados. Dentre os sinais e sintomas, mais de 95% dos casos e óbitos apresentaram febre, tosse e dispneia, menos de 30% apresentaram comorbidades, a ocorrência de hospitalizações dos casos foi de 96 % em Goiás e 45% no Brasil, enquanto que a hospitalização dos que evoluíram para o óbito foi acima de 96%, menos de 14% dos casos e óbitos tomaram a vacina contra a Influenza (H1N1). Concluiu-se que foi possível conhecer o perfil de casos e óbitos a partir das características sócio demográficas e clínicas durante o período epidêmico da Influenza (H1N1) 2009 em Goiás e no Brasil, foram aprendidas muitas lições que auxiliarão na consolidação de planos de enfrentamento a situações inusitadas de caráter epidêmico e pandêmico e norteará a construção de políticas públicas que fortalecerá o sistema de vigilância da doença, da rede de atenção à saúde, implementação de diagnóstico laboratorial, vacinação massiva e medidas de proteção individual e higiene respiratória.

L'augmentation de la taille des jeux de données est une problématique croissante en épidémiologie. La cohorte CoPanFlu-France (1450 sujets), proposant une étude du risque d'infection par la grippe H1N1pdm comme une combinaison de facteurs très divers en est un exemple. Les méthodes statistiques usuelles (e.g. les régressions) pour explorer des associations sont limitées dans ce contexte. Nous comparons l'apport de méthodes exploratoires data-driven à celui de méthodes hypothesis-driven.Une première approche data-driven a été utilisée, évaluant la capacité à détecter des facteurs de l'infection de deux méthodes de data mining, les forêts aléatoires et les arbres de régression boostés, de la méthodologie " régressions univariées/régression multivariée" et de la régression logistique LASSO, effectuant une sélection des variables importantes. Une approche par simulation a permis d'évaluer les taux de vrais et de faux positifs de ces méthodes. Nous avons ensuite réalisé une étude causale hypothesis-driven du risque d'infection, avec un modèle d'équations structurelles (SEM) à variables latentes, pour étudier des facteurs très divers, leur impact relatif sur l'infection ainsi que leurs relations éventuelles. Cette thèse montre la nécessité de considérer de nouvelles approches statistiques pour l'analyse des grands jeux de données en épidémiologie. Le data mining et le LASSO sont des alternatives crédibles aux outils conventionnels pour la recherche d'associations. Les SEM permettent l'intégration de variables décrivant différentes dimensions et la modélisation explicite de leurs relations, et sont dès lors d'un intérêt majeur dans une étude multidisciplinaire comme CoPanFlu. / The increasing size of datasets is a growing issue in epidemiology. The CoPanFlu-France cohort(1450 subjects), intended to study H1N1 pandemic influenza infection risk as a combination of biolo-gical, environmental, socio-demographic and behavioral factors, and in which hundreds of covariatesare collected for each patient, is a good example. The statistical methods usually employed to exploreassociations have many limits in this context. We compare the contribution of data-driven exploratorymethods, assuming the absence of a priori hypotheses, to hypothesis-driven methods, requiring thedevelopment of preliminary hypotheses.Firstly a data-driven study is presented, assessing the ability to detect influenza infection determi-nants of two data mining methods, the random forests (RF) and the boosted regression trees (BRT), ofthe conventional logistic regression framework (Univariate Followed by Multivariate Logistic Regres-sion - UFMLR) and of the Least Absolute Shrinkage and Selection Operator (LASSO), with penaltyin multivariate logistic regression to achieve a sparse selection of covariates. A simulation approachwas used to estimate the True (TPR) and False (FPR) Positive Rates associated with these methods.Between three and twenty-four determinants of infection were identified, the pre-epidemic antibodytiter being the unique covariate selected with all methods. The mean TPR were the highest for RF(85%) and BRT (80%), followed by the LASSO (up to 78%), while the UFMLR methodology wasinefficient (below 50%). A slight increase of alpha risk (mean FPR up to 9%) was observed for logisticregression-based models, LASSO included, while the mean FPR was 4% for the data-mining methods.Secondly, we propose a hypothesis-driven causal analysis of the infection risk, with a structural-equation model (SEM). We exploited the SEM specificity of modeling latent variables to study verydiverse factors, their relative impact on the infection, as well as their eventual relationships. Only thelatent variables describing host susceptibility (modeled by the pre-epidemic antibody titer) and com-pliance with preventive behaviors were directly associated with infection. The behavioral factors des-cribing risk perception and preventive measures perception positively influenced compliance with pre-ventive behaviors. The intensity (number and duration) of social contacts was not associated with theinfection.This thesis shows the necessity of considering novel statistical approaches for the analysis of largedatasets in epidemiology. Data mining and LASSO are credible alternatives to the tools generally usedto explore associations with a high number of variables. SEM allows the integration of variables des-cribing diverse dimensions and the explicit modeling of their relationships ; these models are thereforeof major interest in a multidisciplinary study as CoPanFlu.