Longitudinal analysis of tumor marker CEA of breast cancer patients from Braga's hospital

Allied to an epidemiological study of population of the Senology Unit of Braga’s Hospital that have been diagnosed with malignant breast cancer, we describe the progression in time of repeated measurements of tumor marker Carcinoembryonic antigen (CEA). Our main purpose is to describe the progression of this tumor marker as a function of possible risk factors and, hence, to understand how these risk factors influences that progression. The response variable, values of CEA, was analyzed making use of longitudinal models, testing for different correlation structures. The same covariates used in a previous survival analysis were considered in the longitudinal model. The reference time used was time from diagnose until death from breast cancer. For diagnostic of the models fitted we have used empirical and theoretical variograms. To evaluate the fixed term of the longitudinal model we have tested for a changing point on the effect of time on the tumor marker progression. A longitudinal model was also fitted only to the subset of patients that died from breast cancer, using the reference time as time from date of death until blood test.

Joint modelling of longitudinal and survival data on breast cancer

Tese de Doutoramento em Ciências (Especialidade em Matemática)

Longitudinal analysis of tumor marker CEA of breast cancer patients from Braga's hospital

Allied to an epidemiological study of population of the Senology Unit of Braga’s Hospital that have been diagnosed with malignant breast cancer, we describe the progression in time of repeated measurements of tumor marker Carcinoembryonic antigen (CEA). Our main purpose is to describe the progression of this tumor marker as a function of possible risk factors and, hence, to understand how these risk factors influences that progression. The response variable, values of CEA, was analyzed making use of longitudinal models, testing for different correlation structures. The same covariates used in a previous survival analysis were considered in the longitudinal model. The reference time used was time from diagnose until death from breast cancer. For diagnostic of the models fitted we have used empirical and theoretical variograms. To evaluate the fixed term of the longitudinal model we have tested for a changing point on the effect of time on the tumor marker progression. A longitudinal model was also fitted only to the subset of patients that died from breast cancer, using the reference time as time from date of death until blood test.

Predictors of Weight Change in Sedentary Smokers Receiving a Standard Smoking Cessation Intervention.

INTRODUCTION: Quitting smoking is associated with weight gain, which may threaten motivation to engage or sustain a quit attempt. The pattern of weight gained by smokers treated according to smoking cessation guidelines has been poorly described. We aimed to determine the weight gained after smoking cessation and its predictors, by smokers receiving individual counseling and nicotine replacement therapies for smoking cessation. METHODS: We performed an ancillary analysis of a randomized controlled trial assessing moderate physical activity as an aid for smoking cessation in addition to standard treatment in sedentary adult smokers. We used mixed longitudinal models to describe the evolution of weight over time, thus allowing us to take every participant into account. We also fitted a model to assess the effect of smoking status and reported use of nicotine replacement therapy at each time point. We adjusted for intervention group, sex, age, nicotine dependence, and education. RESULTS: In the whole cohort, weight increased in the first 3 months, and stabilized afterwards. Mean 1-year weight gain was 3.3kg for women and 3.9kg for men (p = .002). Higher nicotine dependence and male sex were associated with more weight gained during abstinence. Age over median was associated with continuing weight gain during relapse. There was a nonsignificant trend toward slower weight gain with use of nicotine replacement therapies. CONCLUSION: Sedentary smokers receiving a standard smoking cessation intervention experience a moderate weight gain, limited to the first 3 months. Older age, male sex, and higher nicotine dependence are predictors of weight gain.

Bromelain as an adjunctive treatment for moderate-to-severe osteoarthritis of the knee: a randomized placebo-controlled pilot study

Background: Osteoarthritis (OA) of the knee is the most prevalent joint disorder. Previous studies suggest that bromelain, a pineapple extract, may be a safer alternative/adjunctive treatment for knee OA than current conventional treatment. Aim: To assess the efficacy of bromelain in treating OA of the knee. Design: Randomized, double-blind placebo-controlled trial. Methods: Subjects (n=47) with a confirmed diagnosis of moderate to severe knee OA were randomized to 12 weeks of bromelain 800 mg/day or placebo, with a 4-week follow-up. Knee (pain, stiffness and function) and quality-of-life symptoms were reported monthly in the WOMAC and SF36 questionnaires, respectively. Adverse events were also recorded. The primary outcome measure was the change in total WOMAC score from baseline to the end of treatment at week 12. Longitudinal models were used to evaluate outcome. Results: Thirty-one patients completed the trial (14 bromelain, 17 placebo). No statistically significant differences were observed between groups for the primary outcome (coefficient 11.16, p=0.27, 95%CI-8.86 to 31.18), nor the WOMAC subscales or SF36. Both treatment groups showed clinically relevant improvement in the WOMAC disability subscale only. Adverse events were generally mild in nature. Discussion: This study suggests that bromelain is not efficacious as an adjunctive treatment of moderate to severe OA, but its limitations support the need for a follow-up study.

The effects of gender segregation at the establishment level: an empirical analysis using a panel of matched employer-employee data

This paper studies the relationship between segregation of women across establish- ments and the wages of males and females. To investigate this issue empirically we use a panel of matched employer-employee data from Brazil. Various longitudinal models are used to assess the wage impact of establishment gender segregation. Overall, the results indicate that the e ect of establishment female proportion on the wages of males and females is negative. We also compare these longitudinal results with cross-section estimates, which are the usual ones obtained in the related literature. This com- parison suggests that unmeasured, time-invariant worker- and establishment-speci c e ects are correlated with the establishment female composition.

Frequency of gynecologic follow-up and cervical cancer screening in the Swiss HIV cohort study

BACKGROUND: According to current recommendations, HIV-infected women should have at least 1 gynecologic examination per year. OBJECTIVES: To analyze factors associated with frequency of gynecologic follow-up and cervical cancer screening among HIV-infected women followed in the Swiss HIV Cohort Study (SHCS). METHODS: Half-yearly questionnaires between April 2001 and December 2004. At every follow-up visit, the women were asked if they had had a gynecologic examination and a cervical smear since their last visit. Longitudinal models were fitted with these variables as outcomes. RESULTS: A total of 2186 women were included in the analysis. Of the 1146 women with complete follow-up in the SHCS, 35.3% had a gynecologic examination in each time period, whereas 7.4% had never gone to a gynecologist. Factors associated with a poor gynecologic follow-up were older age, nonwhite ethnicity, less education, underweight, obesity, being sexually inactive, intravenous drug use, smoking, having a private infectious disease specialist as a care provider, HIV viral load <400 copies/mL, and no previous cervical dysplasia. No association was seen for living alone, CD4 cell count, and positive serology for syphilis. CONCLUSIONS: Gynecologic care among well-followed HIV-positive women is poor and needs to be improved.

O monitoramento domiciliar das condições crônicas e a tomada de decisão por idosos diabéticos e hipertensos

Introdução: A prevalência de doenças crônicas, sobretudo na população idosa, nos coloca diante da necessidade de modelos longitudinais de cuidado. Atualmente os sujeitos estão sendo cada vez mais responsabilizados pelo gerenciamento de sua saúde através do uso de dispositivos de monitoramento, tais como o glicosímetro e o aferidor de pressão arterial. Esta nova realidade culmina na tomada de decisão no próprio domicílio. Objetivos: Identificar a tomada de decisão de idosos no monitoramento domiciliar das condições crônicas; identificar se as variáveis: sexo, escolaridade e renda influenciam a tomada de decisão; identificar a percepção dos idosos quanto às ações de cuidado no domicílio; identificar as dificuldades e estratégias no manuseio dos dispositivos de monitoramento. Materiais e métodos: Estudo quantitativo, exploratório e transversal. Casuística: 150 sujeitos com 60 anos de idade ou mais, sem comprometimento cognitivo, sem depressão e que façam uso do glicosímetro e/ou do aferidor de pressão arterial no domicílio. Instrumentos para seleção dos participantes: (1) Mini Exame do Estado Mental; (2) Escala de Depressão Geriátrica e (3) Escala de Atividades Instrumentais de Vida Diária de Lawton e Brody; Coleta de dados: realizada na cidade de Ribeirão Preto - SP entre setembro de 2014 e outubro de 2015. Instrumentos: (1) Questionário Socioeconômico; (2) Questionário sobre a tomada de decisão no monitoramento da saúde no domicílio (3) Classificação do uso de dispositivos eletrônicos voltados aos cuidados à saúde. Análise dos dados: Realizada estatística descritiva e quantificações absolutas e percentuais para identificar a relação entre tomada de decisão de acordo com o sexo, escolaridade e renda. Resultados: Participaram 150 idosos, sendo 117 mulheres e 33 homens, com média de idade de 72 anos. Destes, 113 são hipertensos e 62 são diabéticos. Quanto à tomada de decisão imediata, tanto os que fazem uso do aferidor de pressão arterial (n=128) quanto do glicosímetro (n=62) referem em sua maioria procurar ajuda médica, seguida da administração do medicamento prescrito e opções alternativas de tratamento. Em médio prazo destaca-se a procura por ajuda profissional para a maioria dos idosos em ambos os grupos. Foi notada pequena diferença na tomada de decisão com relação ao sexo. Quanto à escolaridade, os idosos com mais anos de estudos tendem a procurar mais pelo serviço de saúde se comparado aos idosos de menor escolaridade. A renda não mostrou influencia entre os usuários do glicosímetro. Já entre os usuários do aferidor de pressão arterial, idosos de maior renda tendem a procurar mais pelo serviço de saúde. A maioria dos participantes se refere ao monitoramento domiciliar da saúde de maneira positiva, principalmente pela praticidade em não sair de casa, obtenção rápida de resultados e possibilidade de controle contínuo da doença. As principais dificuldades no manuseio do glicosímetro estão relacionadas ao uso da lanceta e fita reagente, seguida da checagem dos resultados armazenados. Já as dificuldades no uso do aferidor de pressão arterial estão relacionadas a conferir o resultado após cada medida e ao posicionamento correto do corpo durante o monitoramento. Em ambos os grupos as estratégias utilizadas são pedir o auxílio de terceiros e tentativa e erro. Conclusão: Os idosos tem se mostrado favoráveis às ações de monitoramento domiciliar da saúde. De maneira geral, de imediato decidem por ações dentro do próprio domicílio para o controle dos sintomas e isto reforça a necessidade do investimento em informação de qualidade e educação em saúde para que o gerenciamento domiciliar possa vir a ser uma vertente do cuidado integral no tratamento das condições crônicas.

Censored Linear Regression Models For Irregularly Observed Longitudinal Data Using The Multivariate-t Distribution.

In acquired immunodeficiency syndrome (AIDS) studies it is quite common to observe viral load measurements collected irregularly over time. Moreover, these measurements can be subjected to some upper and/or lower detection limits depending on the quantification assays. A complication arises when these continuous repeated measures have a heavy-tailed behavior. For such data structures, we propose a robust structure for a censored linear model based on the multivariate Student's t-distribution. To compensate for the autocorrelation existing among irregularly observed measures, a damped exponential correlation structure is employed. An efficient expectation maximization type algorithm is developed for computing the maximum likelihood estimates, obtaining as a by-product the standard errors of the fixed effects and the log-likelihood function. The proposed algorithm uses closed-form expressions at the E-step that rely on formulas for the mean and variance of a truncated multivariate Student's t-distribution. The methodology is illustrated through an application to an Human Immunodeficiency Virus-AIDS (HIV-AIDS) study and several simulation studies.

Formulating Mixed Models for Experiments, Including Longitudinal Experiments

Mixed models have become important in analyzing the results of experiments, particularly those that require more complicated models (e.g., those that involve longitudinal data). This article describes a method for deriving the terms in a mixed model. Our approach extends an earlier method by Brien and Bailey to explicitly identify terms for which autocorrelation and smooth trend arising from longitudinal observations need to be incorporated in the model. At the same time we retain the principle that the model used should include, at least, all the terms that are justified by the randomization. This is done by dividing the factors into sets, called tiers, based on the randomization and determining the crossing and nesting relationships between factors. The method is applied to formulate mixed models for a wide range of examples. We also describe the mixed model analysis of data from a three-phase experiment to investigate the effect of time of refinement on Eucalyptus pulp from four different sources. Cubic smoothing splines are used to describe differences in the trend over time and unstructured covariance matrices between times are found to be necessary.

Modelos Mixtos-Aditivos Generalizados para el Estudio de Fenómenos Biológicos y Sociales Descriptos por Datos Agregados y/o Longitudinales. Generalized Mixed Additive Models for the Study of Biological and Social Phenomenons from Longitudinal and/or Aggregated Data.

Este proyecto propone extender y generalizar los procesos de estimación e inferencia de modelos aditivos generalizados multivariados para variables aleatorias no gaussianas, que describen comportamientos de fenómenos biológicos y sociales y cuyas representaciones originan series longitudinales y datos agregados (clusters). Se genera teniendo como objeto para las aplicaciones inmediatas, el desarrollo de metodología de modelación para la comprensión de procesos biológicos, ambientales y sociales de las áreas de Salud y las Ciencias Sociales, la condicionan la presencia de fenómenos específicos, como el de las enfermedades.Es así que el plan que se propone intenta estrechar la relación entre la Matemática Aplicada, desde un enfoque bajo incertidumbre y las Ciencias Biológicas y Sociales, en general, generando nuevas herramientas para poder analizar y explicar muchos problemas sobre los cuales tienen cada vez mas información experimental y/o observacional.Se propone, en forma secuencial, comenzando por variables aleatorias discretas (Yi, con función de varianza menor que una potencia par del valor esperado E(Y)) generar una clase unificada de modelos aditivos (paramétricos y no paramétricos) generalizados, la cual contenga como casos particulares a los modelos lineales generalizados, no lineales generalizados, los aditivos generalizados, los de media marginales generalizados (enfoques GEE1 -Liang y Zeger, 1986- y GEE2 -Zhao y Prentice, 1990; Zeger y Qaqish, 1992; Yan y Fine, 2004), iniciando una conexión con los modelos lineales mixtos generalizados para variables latentes (GLLAMM, Skrondal y Rabe-Hesketh, 2004), partiendo de estructuras de datos correlacionados. Esto permitirá definir distribuciones condicionales de las respuestas, dadas las covariables y las variables latentes y estimar ecuaciones estructurales para las VL, incluyendo regresiones de VL sobre las covariables y regresiones de VL sobre otras VL y modelos específicos para considerar jerarquías de variación ya reconocidas. Cómo definir modelos que consideren estructuras espaciales o temporales, de manera tal que permitan la presencia de factores jerárquicos, fijos o aleatorios, medidos con error como es el caso de las situaciones que se presentan en las Ciencias Sociales y en Epidemiología, es un desafío a nivel estadístico. Se proyecta esa forma secuencial para la construcción de metodología tanto de estimación como de inferencia, comenzando con variables aleatorias Poisson y Bernoulli, incluyendo los existentes MLG, hasta los actuales modelos generalizados jerárquicos, conextando con los GLLAMM, partiendo de estructuras de datos correlacionados. Esta familia de modelos se generará para estructuras de variables/vectores, covariables y componentes aleatorios jerárquicos que describan fenómenos de las Ciencias Sociales y la Epidemiología.

In Vivo Longitudinal (1)H MRS Study of Transgenic Mouse Models of Prion Disease in the Hippocampus and Cerebellum at 14.1 T.

In vivo (1)H MR spectroscopy allows the non invasive characterization of brain metabolites and it has been used for studying brain metabolic changes in a wide range of neurodegenerative diseases. The prion diseases form a group of fatal neurodegenerative diseases, also described as transmissible spongiform encephalopathies. The mechanism by which prions elicit brain damage remains unclear and therefore different transgenic mouse models of prion disease were created. We performed an in vivo longitudinal (1)H MR spectroscopy study at 14.1 T with the aim to measure the neurochemical profile of Prnp -/- and PrPΔ32-121 mice in the hippocampus and cerebellum. Using high-field MR spectroscopy we were able to analyze in details the in vivo brain metabolites in Prnp -/- and PrPΔ32-121 mice. An increase of myo-inositol, glutamate and lactate concentrations with a decrease of N-acetylaspartate concentrations were observed providing additional information to the previous measurements.

Bayesian longitudinal data analysis with mixed models and thick-tailed distributions using MCMC

Linear mixed effects models are frequently used to analyse longitudinal data, due to their flexibility in modelling the covariance structure between and within observations. Further, it is easy to deal with unbalanced data, either with respect to the number of observations per subject or per time period, and with varying time intervals between observations. In most applications of mixed models to biological sciences, a normal distribution is assumed both for the random effects and for the residuals. This, however, makes inferences vulnerable to the presence of outliers. Here, linear mixed models employing thick-tailed distributions for robust inferences in longitudinal data analysis are described. Specific distributions discussed include the Student-t, the slash and the contaminated normal. A Bayesian framework is adopted, and the Gibbs sampler and the Metropolis-Hastings algorithms are used to carry out the posterior analyses. An example with data on orthodontic distance growth in children is discussed to illustrate the methodology. Analyses based on either the Student-t distribution or on the usual Gaussian assumption are contrasted. The thick-tailed distributions provide an appealing robust alternative to the Gaussian process for modelling distributions of the random effects and of residuals in linear mixed models, and the MCMC implementation allows the computations to be performed in a flexible manner.

Competing regression models for longitudinal data

The choice of an appropriate family of linear models for the analysis of longitudinal data is often a matter of concern for practitioners. To attenuate such difficulties, we discuss some issues that emerge when analyzing this type of data via a practical example involving pretestposttest longitudinal data. In particular, we consider log-normal linear mixed models (LNLMM), generalized linear mixed models (GLMM), and models based on generalized estimating equations (GEE). We show how some special features of the data, like a nonconstant coefficient of variation, may be handled in the three approaches and evaluate their performance with respect to the magnitude of standard errors of interpretable and comparable parameters. We also show how different diagnostic tools may be employed to identify outliers and comment on available software. We conclude by noting that the results are similar, but that GEE-based models may be preferable when the goal is to compare the marginal expected responses.