Diss. (sammanfattning) Stockholm : Karol. inst., 2003. / Härtill 4 uppsatser.

Diss. (sammanfattning) Stockholm : Karol. inst., 2004. / Härtill 5 uppsatser.

Thesis (Ph. D.)--University of Missouri-Columbia, 1996. / Typescript. Vita. Includes bibliographical references (leaves 105-108). Also available on the Internet.

Thesis (Ph. D.)--University of Missouri-Columbia, 1996. / Typescript. Vita. Includes bibliographical references (leaves 105-108). Also available on the Internet.

Thesis (Ph. D.)--Case Western Reserve University, 2006. / [School of Medicine] Department of Epidemiology and Biostatistics. Includes bibliographical references. Available online via OhioLINK's ETD Center.

Thesis (M.A.)--University of South Florida, 2006. / Title from PDF of title page. Document formatted into pages; contains 71 pages. Includes bibliographical references.

Made available in DSpace on 2015-03-26T13:32:05Z (GMT). No. of bitstreams: 1 texto completo.pdf: 557235 bytes, checksum: 0ad92dee7beb22112fd8ec7de711614d (MD5) Previous issue date: 2009-02-19 / Fundação de Amparo a Pesquisa do Estado de Minas Gerais / The aim of this study was to compare non-linear models fitted to data on survival of ants under different treatments using alternative methodology. This methodology consists of using the technique of cluster analysis, Ward's method, to evaluate the identity of models used in survival analysis. The data used in this work are related to an experiment conducted in the entomology laboratory of the Federal University of Viçosa, Brazil. It also used a set of simulated data based on the Weibull distribution. Originally was applied a nonparametric technique, Kaplan- Meyer estimator, to estimate the survival curves, then the logrank test for comparison of these curves. For the real dataset it was fitted a logistic model of survival times, while for the simulated data it was fitted a Weibull model. The estimates of the parameters of each fitted model were grouped using the technique of cluster analysis. The results found by the grouping were equivalent to that by the logrank test. It is concluded that the proposed methodology showed to be efficient and less laborious, mainly when many survival curves need to be compared. / O objetivo desse trabalho foi comparar modelos não-lineares ajustados aos dados de sobrevivência de formigas submetidas a diferentes tratamentos através de metodologia alternativa. Essa metodologia consistiu no uso da técnica de análise de agrupamento, método de Ward, para a identidade de modelos usados em análise de sobrevivência. Os dados utilizados neste trabalho são referentes a um experimento realizado no laboratório de entomologia da Universidade Federal de Viçosa. Foi também utilizado um conjunto de dados simulado com base na distribuição de Weibull. Inicialmente aplicou-se técnica não paramétrica, estimador Kaplan-Meier, a fim de estimar as curvas de sobrevivência de cada tratamento e, em seguida, o teste logrank para a comparação dessas curvas. Para os dados reais foi ajustado o modelo logístico aos tempos de sobrevivência, enquanto que, para os dados simulados foi ajustado o modelo de Weibull. Para cada caso agrupou-se os parâmetros estimados de cada modelo utilizando-se as técnicas de análise de agrupamento. Os resultados encontrados pelo agrupamento foram equivalentes aos do teste logrank. Concluiu-se que a metodologia proposta mostrou ser eficiente e menos trabalhosa, quando várias curvas de sobrevivência precisam ser comparadas.

Regression models in survival analysis and reliability Doctoral thesis Petr Novák Charles University in Prague, Faculty of Mathematics and Physics Abstract: In present work we study methods for modeling the dependence of data from sur- vival and reliability setting on available explanatory variables. The first part of the work compares the properties of the Cox proportional hazards model, Aalen additive model and the Accelerated failure model for survival data. We present methods for testing goodness-of-fit based on counting processes and martingale theory, allowing to identify which model fits the data best. The second part focuses on modeling the lifetime of repairable systems. We study the means of incorporating the history of studied devices into the models, including the influence of corrective repairs and preventive maintenance actions. We demonstrate the introduced methods on real applications and study their properties in various situations on simulated data. 1

Orientador: Sandra Aidar de Queiroz / Coorientador: Sabrina Luzia Caetano / Banca: Janaina Della Torre da Silva / Banca: Patricia Tholon / Banca: Danísio Prado Munari / Banca: Isabel Cristina Boreli / Resumo: O estudo dos comportamentos de imobilidade tônica (IT) e de reintegração social (CRS) pode revelar as interações homem-animal e animalanimal, refletindo no bem-estar dos animais no ambiente de cativeiro, acarretando, assim, em respostas positivas ou negativas para a produção e lucratividade do sistema de criação. Visando entender o nível de medo e a manifestação de sociabilidade em perdizes (Rhynchotus rufescens), a metodologia de análise de sobrevivência foi utilizada para estimar o tempo de permanência em IT e o tempo transcorrido para um animal se unir com seu conspecífico. Para isto, perdizes criadas no Setor de Animais Silvestres - FCAV - Unesp submetidas às avaliações de IT e de CRS, no período de 2006 a 2010, foram utilizadas. As observações de IT foram realizadas em 539 aves e os resultados revelaram que os efeitos de ano de nascimento, mês e peso corporal da ave na época da medida de IT foram importantes fontes de variação na duração do tempo em catatonia. Além disso, aves com maior peso corporal tenderam a permanecer maior tempo em IT. Observações de CRS realizados em 503 aves mostraram que as mais velhas levaram mais tempo para unir-se ao conspecífico. O efeito aleatório de pai foi significativo (p < 0,05) para IT e a herdabilidade estimada (h2) empregando-se o software Survival Kit foi igual a 0,37. A h2 para CRS foi 0,31, entretanto supõe-se alto erro-padrão associado a esta estimativa devido ao efeito aleatório de pai não ter sido significativo (p < 0,05). A correlação entre os valores genéticos de IT e de CRS foi 0,0598 (p = 0,1491), evidenciando a inexistência de genes de efeito aditivo comuns entre estes dois comportamentos na população analisada. É possível aplicar a seleção nas aves para diminuir o tempo de permanência em IT, sem alterar o nível de socialização / Abstract: The study of tonic immobility (TI) and social reinstatement behavior (SRB) can reveal interactions between man-animal and animal-animal. Those interactions reflect on the welfare of captive animals and may result in positively or negatively effects on the production and profitability of production system. Aiming at understanding the level of fear and the manifestation of sociability in red-winged tinamou (Rhynchotus rufescens), the survival analysis was applied to estimate the time in TI and also the time expent to an animal join with its conspecific. For this, redwinged tinamous reared in the Setor de Animais Silvestres - FCAV - Unesp hatched from 2006 to 2010 underwent TI and SRB evaluations. The observations of TI were performed in 539 birds and the results revealed that the effects of year of birth, month, and body weight of the bird at the time the measure of TI were important sources of variation for the time in catatonia; furthermore, birds with higher body weight tended to remain longer in TI. The SRB observations made in 503 birds showed that the older birds spent more time then the younger to join with their conspecific. The random effect of sire was significant (p < 0.05) for TI and the estimated heritability (h2) by Survival Kit software was equal to 0.37. The h2 for SRB was 0.31, however the estimate is supposed to have a high standard error associated with due to the random effect of sire being not significant. The correlation between the predicted breeding values of TI and SRB was 0.0598 (p = 0.1491), evidencing the non existence of common additive genes between these two traits in the studied population. Therefore, selection to decrease the time in TI could be applied on TI, without changing the level of socialization in this population / Doutor

Objective: This research explored how time-to-event dental articles were indexed and reported, and sought solutions to help improve the reporting and identification of these articles, so that they could be more easily found and used to inform practice and research. Methods: Articles reporting time-to-event dental outcomes in humans were identified from the 50 dental journals with the highest impact factor for 2008. These were handsearched, identifying 'case' articles (n=95), active controls (likely false positives, n=91), and passive controls (other true negatives, n=6796). The medical subject headings (MESH) that had been assigned to the articles in MEDLINE, and words used in titles and abstracts describing time-to-events were compared between the 'cases' and controls. Time-to-event words and figures within articles were also sought, and reporting quality was assessed. Search strategies to identify time-toevent articles were developed, using indexing terms and free-text words. An independent cohort of articles was used to validate the search strategies, consisting of 148 time-to-event articles handsearched from 6514 articles in the 50 dental journals with the highest impact factor for 2012. The findings of the research were used to draft guidance to improve reporting, which was circulated amongst 78 stakeholder experts for comment, and modified. Results: MeSH indexing of time-to-event analyses was inconsistent and inaccurate, author descriptions in abstracts and titles varied, and the quality of time-to-event reporting and graphics in the body of those articles was poor. The burden faced by someone wishing to find and use these articles was considered high. Sensitive, precise and optimized electronic search strategies were developed and validated with sensitivities up to 92% and precisions up to 93%. The draft guidance attracted comment from 46 experts across 15 countries, with approximately 90% of the 130 comments accepted into the revised version. The importance of good quality reporting was endorsed, and there was high interest in commending the guidance to authors, reviewers, and training dental specialists. Conclusions: This research programme explored how time-to-event dental articles were reported, and used those findings to suggest solutions that would help to improve the identification and use of these data, reducing research waste.