Thesis (Ph.D. in Biostatistics) -- University of Colorado Denver, 2008. / Typescript. Includes bibliographical references (leaves 117-119). Free to UCD Anschutz Medical Campus. Online version available via ProQuest Digital Dissertations;

No description available.

INTRODUÇÃO: O status axilar é o fator prognóstico mais importante em câncer de mama. A biópsia de linfonodo sentinela (BLS) tornou-se o procedimento padrão no estadiamento axilar em pacientes com axila clinicamente negativa. O procedimento recomendado para pacientes com metástase em linfonodo sentinela (LS) inclui a linfadenectomia axilar completa (LAC). Porém questiona-se a necessidade da LAC em todos os pacientes com metástase em LS (MLS), particularmente naqueles com baixo risco de metástase em linfonodos axilares adicionais (Não-LS). Estimar de forma precisa a probabilidade de MLS e metástase em Não-LS (MNão-LS) pode auxiliar em muito o processo de decisão terapêutica. Os dados publicados referentes aos fatores preditivos de MLS e MNão-LS são de certa forma escassos. A forma de como esses dados são apresentados – geralmente expressos como razão de chance (odds-ratio) – dificulta o cálculo de probabilidade de MLS ou MNão-LS para um paciente em específico. Nesta tese, dois modelos de previsão computadorizados de fácil utilização foram desenvolvidos, utilizando-se um grande número de casos, para facilitar o cálculo de probabilidade de MLS e MNão-LS. MÉTODOS: Todos os dados clínico-patológicos foram coletados do banco de dados prospectivo de LS do Memorial Sloan-Kettering Cancer Center (MSKCC), Nova York, EUA. Os projetos de desenvolvimento de ambos os modelos foram aprovados pelo Institutional Review Board do MSKCC. Dois modelos foram desenvolvidos: Modelo de MLS (Modelo LS) e Modelo de metástase em Não-LS (Modelo Não-LS). No Modelo LS, os achados clínico-patológicos de 4.115 procedimentos subseqüentes de BLS (amostra de modelagem) foram submetidos à análise de regressão logística multivariada para se criar um modelo de previsão de MLS. Um software baseado nesse modelo foi desenvolvido utilizando-se as variáveis: idade, tamanho do tumor, tipo histológico, invasão vásculo-linfática (IVL), localização do tumor e multifocalidade. Esse modelo foi validado em uma amostra distinta (amostra de validação) com 1.792 BLSs subseqüentes. No Modelo Não-LS, os achados patológicos do tumor primário e das MLS, obtidas de 702 procedimentos de BLS (amostra de modelagem) em pacientes submetidos à LAC, foram submetidos à análise de regressão logística multivariada para se criar um modelo de previsão de MNão-LS. Um nomograma e um software baseados nesse modelo foram desenvolvidos utilizando-se as variáveis: tamanho do tumor, tipo histológico, grau nuclear, IVL, multifocalidade, receptor de estrógeno, método de detecção da MLS, número de LS positivos, número de LS negativos. Esse modelo foi validado em uma amostra distinta (amostra de validação) com 373 casos subseqüentes. RESULTADOS: O software do Modelo LS na amostra de modelagem mostrou-se adequado, com a área sob a curva de características operacionais (ROC) de 0,76. Quando aplicado na amostra de validação, o Modelo LS também previu de forma acurada as probabilidades de MLS (ROC = 0,76). O software e o nomograma do Modelo Não-LS na amostra de modelagem apresentaram uma área ob a curva ROC de 0,76 e, na amostra de validação, 0,77. CONCLUSÕES: Dois softwares de fácil utilização foram desenvolvidos, utilizando-se informações comumente disponíveis pelo cirurgião para calcular para o paciente a probabilidade de MLS e MNão-LS de forma precisa, fácil e individualizada. O software do Modelo LS não deve ser utilizado, porém, para se evitar a BLS. Para download dos softawares clique em: <A HREF="http://www.mastologia.com" TARGET="_BLANK">www.mastologia.com . / INTRODUCTION: Axillary lymph node status is the most significant prognostic factor in breast cancer. Sentinel lymph node biopsy (SLNB) has become the standard of care as the axillary staging procedure in clinically node negative patients. The standard procedure for patients with sentinel lymph node (SLN) metastasis includes complete axillary lymph node dissection (ALND). However, many experts question the need for complete ALND in every patient with detectable SLN metastases, particularly those perceived to have a low risk of additional lymph node (Non-SLN) metastasis. Accurate estimates of the likelihood of SLN metastases and additional disease in the axilla could greatly assist in decision-making treatment. The published data on predictive factors for SLN and Non-SLN metastases is somewhat scarce. It is also difficult to apply these data – usually expressed as odds ratio – to calculate the probability of SLN or Non-SLN metastases for a specific patient. In this thesis, two user-friendly computerized prediction models based on large datasets were developed, to assist the prediction of the presence of SLN and Non-SLN metastases. METHODS: All clinical and pathological data were collected from the prospective SLN database of Memorial Sloan-Kettering Cancer Center (MSKCC), New York, USA. The development projects of both models were approved by the Institutional Review Board of MSKCC. Two models were developed: Model for predicting SLN metastases (SLN Model) and Model for predicting Non-SLN metastases (Non-SLN Model). In the SLN Model, clinical and pathological features of 4,115 sequential SLNB procedures (modeling sample) were assessed with multivariable logistic regression to predict the presence SLN metastases. A software based on the logistic regression model was created using age, tumor size, tumor type, lymphovascular invasion, tumor location and multifocality. This model was subsequently applied to another set of 1,792 sequential SLNBs (validation sample). In the Non-SLN Model, pathological features of the primary tumor and SLN metastases, identified in 702 SLNBs (modeling sample) on patients who underwent complete ALND, were assessed with multivariable logistic regression to predict the presence of additional disease in the Non-SLNs of these patients. A nomogram and a software were created using tumor size, tumor type and nuclear grade, lymphovascular invasion, multifocality, and estrogen-receptor status of the primary tumor; method of detection of SLN metastases; number of positive SLNs; and number of negative SLNs. This model was subsequently applied to another set of sequential 373 procedures (validation sample). RESULTS: The software of the SLN Model for the modeling sample was accurate and discriminating, with an area under the receiver operating characteristic (ROC) curve of 0.76. When applied to the validation sample, the SLN-Model accurately predicted likelihood of SLN metastases (ROC = 0.76). The software and nomogram of the Non-SLN Model was also accurately predicted likelihood of Non-SLN metastases, with an area under ROC curve of 0.76 for the modeling sample, and 0.77 for the validation sample. CONCLUSION: Two user-friendly softwares were developed, using information commonly available to the surgeon to easily and accurately calculate the likelihood of having SLN metastases or additional, Non-SLN metastases for individual patients. However, the software of the SLN Model should not be used to avoid SLNB. Click on <A HREF="http://www.mastologia.com" TARGET="_BLANK">www.mastologia.com to download these softwares.

Background. The randomized controlled clinical trial is currently the most scientific method for evaluating the effect of medical interventions. The sample size of a trial is crucial for reliably estimating the effect. However, many clinical trials may not be sufficiently large in size to detect the effect of interventions assessed. Previous studies of the statistical power, a relative measure of the largeness of a study, were normally small, mainly examined trials with a statistically insignificant result and were flawed because of the biased or purely hypothetical estimate of the effect for the computation of the power. By using meta-analysis, we conducted this study with improved methods for estimating the power and included a larger number of trials. / Findings. A total of 2,923,912 patients from 2,872 clinical trials from 466 systematic reviews were included in the analyses of this thesis. Of the 466 systematic reviews, 24% (113) were identified from the five journals and the remaining 76% (353) were from the Cochrane Library. 1,000 trials and 1,583,204 patients were obtained from 113 systematic reviews identified in the journals, in which 13.7% (95% C.I.: 11.6%, 15.8%) of trials had a sufficient power and the overall power was 34.0% (95% C.I.: 33.7%, 34.3%). 1,872 trials and 1,340,708 patients were obtained from 353 systematic reviews identified in the Cochrane Library, in which 16.7% (95% C.I.: 15.0%, 18.4%) of trials had a sufficient power and the overall power was 37.8% (95% C.I.: 37.6%, 38.0%). (Abstract shortened by UMI.) / Methods. We identified trials from systematic reviews of clinical trials with binary outcomes published in five medical journals and the Cochrane Database of Systematic Reviews. We analyzed the power of trials with a significant result as well as those with an insignificant result. In estimating the power, we used the combined odds ratio of the meta-analysis as the estimate of the effect for trials from systematic reviews with a statistically significant result and a relative risk reduction of 25% for trials from systematic reviews with a statistically insignificant result. In addition to use of the conventional method to estimate the power, we also developed a new "counting method" that does not need any assumption about the effect. Furthermore, the power is also expressed as a relative and absolute difference between the number of subjects required for a power of 80% and that actually recruited by the trials. / Tsoi Kam Fai. / "July 2005." / Adviser: Jin Ling Tang. / Source: Dissertation Abstracts International, Volume: 67-01, Section: B, page: 0161. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (p. 107-113). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Submitted by Rosivalda Pereira (mrs.pereira@ufma.br) on 2017-06-22T20:19:45Z No. of bitstreams: 1 RainaLima.pdf: 6488864 bytes, checksum: db557811216cf129eb2f4baa56f2046f (MD5) / Made available in DSpace on 2017-06-22T20:19:45Z (GMT). No. of bitstreams: 1 RainaLima.pdf: 6488864 bytes, checksum: db557811216cf129eb2f4baa56f2046f (MD5) Previous issue date: 2016-02-18 / Objective. This study aims to analyze the effects of prepregnancy nutritional status and gestational weight gain on birth weight. Methodology. Cross-sectional study involving 5,024 mothers and their newborns who participated in the study BRISA São Luís - MA. Data were collected in 2010 and were applied two questionnaires after delivery: one with maternal data and other newborn´s data. The main explanatory variables were prepregnancy BMI and gestational weight gain. Theoretical model was proposed to explain the total, direct and indirect effects, using structural equation modeling in the analysis, with adjustment for sociodemographic variables, life habits and maternal comorbidities, with the outcome birth weight. Results. The final model had good fit according to indicators RMSEA, CFI / TLI and WRMR. The prepregnancy BMI had total effect (Standardized Coefficient SC=0.126; p <0.001) and direct (SC=0.211; p <0.001) positive on the newborn's weight, plus negative indirect effect on the total gestational weight gain. Gestational weight gain in turn had the highest effect on birth weight (SC=0.280; p <0.001), including modifying the effect of other variables. Socioeconomic status, maternal age, more stable marital status and gestational diabetes had positive total effects, while high blood pressure and smoking during pregnancy had negative effects on birth weight. Alcohol use during pregnancy showed no total effect. Conclusion. Mothers with higher prepregnancy BMI can have children with higher weight, as well as those with high gestational weight gain. These associations highlight the need for greater attention to the health of women of reproductive age and maintaining proper weight gain during pregnancy, which could contribute to reducing risks of maternal complications and newborn. / Objetivo. O presente estudo teve por objetivo analisar os efeitos do estado nutricional pré-gestacional e ganho de peso gestacional no peso ao nascer. Metodologia. Estudo transversal que envolveu 5.024 mães e seus recém-nascidos que participaram do estudo BRISA de São Luís – MA. Os dados foram coletados no ano de 2010 e aplicaram-se dois questionários após o parto: um com dados da mãe e outro do recém-nascido. As variáveis explanatórias principais foram o IMC pré-gestacional e o ganho de peso gestacional. Modelo teórico foi proposto para explicar efeitos totais, diretos e indiretos, utilizando modelagem de equações estruturais na análise, com ajuste para variáveis sociodemográficas, hábitos de vida e comorbidades maternas, tendo como desfecho peso ao nascer. Resultados. O modelo final teve bom ajuste segundo os indicadores RMSEA, CFI/TLI e WRMR. O IMC pré-gestacional teve efeitos total (Coeficiente padronizado CP=0.126; p<0.001) e direto (CP=0.211; p<0.001) positivos no peso do recém-nascido, além de efeito indireto negativo via ganho de peso gestacional total. O ganho de peso gestacional por sua vez apresentou o maior efeito no peso de nascimento (CP=0.280; p<0.001), inclusive modificando o efeito de outras variáveis. Situação socioeconômica, idade materna, situação conjugal mais estável e diabetes gestacional tiveram efeitos totais positivos, enquanto que hipertensão arterial e tabagismo na gestação apresentaram efeitos negativos no peso ao nascer. Uso de álcool durante a gestação não apresentou efeito total. Conclusão. Mães com maior IMC pré-gestacional podem gerar filhos com mais alto peso, assim como aquelas com elevado ganho de peso total. Essas associações ressaltam a necessidade de maior atenção à saúde de mulheres em idade reprodutiva e manutenção do ganho de peso adequado durante a gestação, o que poderá contribuir para a diminuição de riscos de intercorrências maternas e do recém-nascido.

Thesis (Ph.D. in Biostatistics) -- University of Colorado at Denver and Health Sciences Center, 2006. / Typescript. Includes bibliographical references (leaves 72-75). Free to UCD Anschutz Medical Campus. Online version available via ProQuest Digital Dissertations;

Thesis (Ph.D. in Analytic Health Sciences, Dept. of Preventive Medicine and Biometrics) -- University of Colorado Denver, 2007. / Typescript. Includes bibliographical references (leaves 57-59). Free to UCD Anschutz Medical Campus. Online version available via ProQuest Digital Dissertations;

Thesis (Ph.D. in Analytic Health Sciences, Department of Preventive Medicine and Biometrics) -- University of Colorado Denver, 2007. / Typescript. Includes bibliographical references (leaves 75-77). Free to UCD affiliates. Online version available via ProQuest Digital Dissertations;

Thesis (Ph.D. in Biostatistics, Department of Preventive Medicine and Biometrics) -- University of Colorado Denver, 2007. / Typescript. Includes bibliographical references (leaves 89-93). Free to UCD affiliates. Online version available via ProQuest Digital Dissertations;

INTRODUÇÃO: O status axilar é o fator prognóstico mais importante em câncer de mama. A biópsia de linfonodo sentinela (BLS) tornou-se o procedimento padrão no estadiamento axilar em pacientes com axila clinicamente negativa. O procedimento recomendado para pacientes com metástase em linfonodo sentinela (LS) inclui a linfadenectomia axilar completa (LAC). Porém questiona-se a necessidade da LAC em todos os pacientes com metástase em LS (MLS), particularmente naqueles com baixo risco de metástase em linfonodos axilares adicionais (Não-LS). Estimar de forma precisa a probabilidade de MLS e metástase em Não-LS (MNão-LS) pode auxiliar em muito o processo de decisão terapêutica. Os dados publicados referentes aos fatores preditivos de MLS e MNão-LS são de certa forma escassos. A forma de como esses dados são apresentados – geralmente expressos como razão de chance (odds-ratio) – dificulta o cálculo de probabilidade de MLS ou MNão-LS para um paciente em específico. Nesta tese, dois modelos de previsão computadorizados de fácil utilização foram desenvolvidos, utilizando-se um grande número de casos, para facilitar o cálculo de probabilidade de MLS e MNão-LS. MÉTODOS: Todos os dados clínico-patológicos foram coletados do banco de dados prospectivo de LS do Memorial Sloan-Kettering Cancer Center (MSKCC), Nova York, EUA. Os projetos de desenvolvimento de ambos os modelos foram aprovados pelo Institutional Review Board do MSKCC. Dois modelos foram desenvolvidos: Modelo de MLS (Modelo LS) e Modelo de metástase em Não-LS (Modelo Não-LS). No Modelo LS, os achados clínico-patológicos de 4.115 procedimentos subseqüentes de BLS (amostra de modelagem) foram submetidos à análise de regressão logística multivariada para se criar um modelo de previsão de MLS. Um software baseado nesse modelo foi desenvolvido utilizando-se as variáveis: idade, tamanho do tumor, tipo histológico, invasão vásculo-linfática (IVL), localização do tumor e multifocalidade. Esse modelo foi validado em uma amostra distinta (amostra de validação) com 1.792 BLSs subseqüentes. No Modelo Não-LS, os achados patológicos do tumor primário e das MLS, obtidas de 702 procedimentos de BLS (amostra de modelagem) em pacientes submetidos à LAC, foram submetidos à análise de regressão logística multivariada para se criar um modelo de previsão de MNão-LS. Um nomograma e um software baseados nesse modelo foram desenvolvidos utilizando-se as variáveis: tamanho do tumor, tipo histológico, grau nuclear, IVL, multifocalidade, receptor de estrógeno, método de detecção da MLS, número de LS positivos, número de LS negativos. Esse modelo foi validado em uma amostra distinta (amostra de validação) com 373 casos subseqüentes. RESULTADOS: O software do Modelo LS na amostra de modelagem mostrou-se adequado, com a área sob a curva de características operacionais (ROC) de 0,76. Quando aplicado na amostra de validação, o Modelo LS também previu de forma acurada as probabilidades de MLS (ROC = 0,76). O software e o nomograma do Modelo Não-LS na amostra de modelagem apresentaram uma área ob a curva ROC de 0,76 e, na amostra de validação, 0,77. CONCLUSÕES: Dois softwares de fácil utilização foram desenvolvidos, utilizando-se informações comumente disponíveis pelo cirurgião para calcular para o paciente a probabilidade de MLS e MNão-LS de forma precisa, fácil e individualizada. O software do Modelo LS não deve ser utilizado, porém, para se evitar a BLS. Para download dos softawares clique em: <A HREF="http://www.mastologia.com" TARGET="_BLANK">www.mastologia.com . / INTRODUCTION: Axillary lymph node status is the most significant prognostic factor in breast cancer. Sentinel lymph node biopsy (SLNB) has become the standard of care as the axillary staging procedure in clinically node negative patients. The standard procedure for patients with sentinel lymph node (SLN) metastasis includes complete axillary lymph node dissection (ALND). However, many experts question the need for complete ALND in every patient with detectable SLN metastases, particularly those perceived to have a low risk of additional lymph node (Non-SLN) metastasis. Accurate estimates of the likelihood of SLN metastases and additional disease in the axilla could greatly assist in decision-making treatment. The published data on predictive factors for SLN and Non-SLN metastases is somewhat scarce. It is also difficult to apply these data – usually expressed as odds ratio – to calculate the probability of SLN or Non-SLN metastases for a specific patient. In this thesis, two user-friendly computerized prediction models based on large datasets were developed, to assist the prediction of the presence of SLN and Non-SLN metastases. METHODS: All clinical and pathological data were collected from the prospective SLN database of Memorial Sloan-Kettering Cancer Center (MSKCC), New York, USA. The development projects of both models were approved by the Institutional Review Board of MSKCC. Two models were developed: Model for predicting SLN metastases (SLN Model) and Model for predicting Non-SLN metastases (Non-SLN Model). In the SLN Model, clinical and pathological features of 4,115 sequential SLNB procedures (modeling sample) were assessed with multivariable logistic regression to predict the presence SLN metastases. A software based on the logistic regression model was created using age, tumor size, tumor type, lymphovascular invasion, tumor location and multifocality. This model was subsequently applied to another set of 1,792 sequential SLNBs (validation sample). In the Non-SLN Model, pathological features of the primary tumor and SLN metastases, identified in 702 SLNBs (modeling sample) on patients who underwent complete ALND, were assessed with multivariable logistic regression to predict the presence of additional disease in the Non-SLNs of these patients. A nomogram and a software were created using tumor size, tumor type and nuclear grade, lymphovascular invasion, multifocality, and estrogen-receptor status of the primary tumor; method of detection of SLN metastases; number of positive SLNs; and number of negative SLNs. This model was subsequently applied to another set of sequential 373 procedures (validation sample). RESULTS: The software of the SLN Model for the modeling sample was accurate and discriminating, with an area under the receiver operating characteristic (ROC) curve of 0.76. When applied to the validation sample, the SLN-Model accurately predicted likelihood of SLN metastases (ROC = 0.76). The software and nomogram of the Non-SLN Model was also accurately predicted likelihood of Non-SLN metastases, with an area under ROC curve of 0.76 for the modeling sample, and 0.77 for the validation sample. CONCLUSION: Two user-friendly softwares were developed, using information commonly available to the surgeon to easily and accurately calculate the likelihood of having SLN metastases or additional, Non-SLN metastases for individual patients. However, the software of the SLN Model should not be used to avoid SLNB. Click on <A HREF="http://www.mastologia.com" TARGET="_BLANK">www.mastologia.com to download these softwares.