Orientadores: Luiz Carlos de Freitas, Dalton Francisco de Andrade / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Educação / Made available in DSpace on 2018-08-24T22:44:15Z (GMT). No. of bitstreams: 1 Dalben_Adilson_D.pdf: 5011742 bytes, checksum: e9c6413b4e6fb98c276dcbdecd13440b (MD5) Previous issue date: 2014 / Resumo: Esta pesquisa é um estudo sobre a eficácia e equidade escolar que tem ganhado atenção especial nos países que usam as avaliações em larga escala a serviço da gestão do sistema educativo. No Brasil, que desde a década de 1990 colocou a avaliação educacional como recurso central em suas políticas educacionais, mas coletando dados seccionais, que são muito frágeis para essa finalidade. Essa fragilidade decorre da alta associação que os fatores extraescolares, sobretudo o nível socioeconômico do aluno, têm sobre as medidas de proficiência. Diante disso, foram usados dados longitudinais e a análise foi feita por meio de modelos lineares hierárquicos. Esta pesquisa teve como objetivo principal desenvolver um modelo estatístico capaz de identificar tais fatores para a realidade brasileira, considerando que a aprendizagem é um processo complexo, isto é, ela é influenciada simultaneamente por múltiplos fatores. Foram desenvolvidos modelos de valor agregado que não só identificam tais variáveis, como também caracterizam sua influência em alunos com distintas proficiências no início de cada período de escolarização. A base de dados utilizada nesses modelos foi fornecida pelo Projeto GERES, que, no período de 2005 a 2008, coletou dados dos mesmos alunos de 1ª a 4ª séries de uma amostra de 312 escolas em cinco grandes cidades brasileiras. Foram medidas as proficiências em Leitura e Matemática de 35.538 alunos e coletadas informações de contexto desses alunos, seus familiares, professores, diretores e escola. Após a redução do grande número de informações disponibilizadas pelo Projeto GERES, feita por meio da Análise Fatorial Exploratória (AFE), as variáveis resultantes foram reorganizadas em três arquivos usados para análise em modelos lineares hierárquicos de três níveis. Os resultados encontrados evidenciam uma significativa instabilidade nos efeitos que as variáveis têm sobre a proficiência, tanto em leitura quanto em matemática. Ao final da pesquisa, são encontrados alguns fatores que influenciam positivamente e negativamente a proficiência em Leitura e Matemática e outros que afetam especificamente cada uma dessas áreas, indicando que podem colaborar para o aumento da eficácia e da equidade das escolas. No entanto, constatam-se também algumas variáveis que têm comportamentos incoerentes com o esperado e outras com comportamentos opostos nas duas áreas. Assim, dos achados das pesquisas, comprova-se que, com base nos dados utilizados, procedimentos metodológicos e modelos estatísticos adotados, os modelos de valor agregado melhoram a confiabilidade das análises em comparação aos modelos que usam dados seccionais, mas ainda são inviáveis como ferramentas para a gestão do sistema educativo, sobretudo para o uso meritocrático de seus resultados. Dessa forma, esta pesquisa corrobora os achados de outras realizadas no âmbito internacional e permite afirmar que a qualidade da modelagem estatística depende da qualidade dos dados que busca modelar, podendo gerar distorções, estabelecer relações inesperadas ou levar a conclusões equivocadas. Em contrapartida, trata-se de recursos que podem ser usados no sistema educativo, fornecendo dados importantes para a orientação das políticas públicas numa perspectiva de avaliação formativa, com vistas ao melhoramento da qualidade de ensino oferecido pelas escolas e à melhor formação dos profissionais docentes e não-docentes que nelas trabalham / Abstract: This research is a study on school effectiveness and equality in Brazil, adding up to a number of other researches that have drawn special attention in countries that use large-scale evaluations at the service of the education system management. In the Brazil has regarded the educational evaluation as a central resource in national education policies, but using cross-sectional data, which are far more fragile for such purpose. This fragility has derived from the great influence that extra-school factors, particularly the students¿ socioeconomic status, exerts on proficiency measures. Longitudinal data was used in the analyses with hierarchical linear models. The main objective of this research was to develop a statistical model to identify such factors in the Brazilian reality, considering that learning is a complex process, i.e. it is simultaneously influenced by multiple factors. Value-added models were developed not only to identify such variables, but also to characterize their influence on students showing different proficiencies at the beginning of every school term. The data base used in those models was provided by the GERES Project, which collected data of the same students from the 1st to the 4th grade from a sample of 312 schools in five Brazilian cities from 2005 to 2008. Proficiencies of 35,538 students were measured, and information about these students¿ context, family, teachers, principals and school were gathered. After the reduction of the great amount of information made available by the GERES Project by means of Exploratory Factor Analysis (EFA), the resulting variables were reorganized in three files used for analysis in three-level hierarchical linear models. The results evidenced significant instability in the effects that the variables have on proficiency both in Reading and in Mathematics. At the end of the research, some factors that influence Reading and Mathematics proficiency either positively or negatively, as well as other factors that specifically affect one of those areas, were found, thus indicating that they may contribute to increased school effectiveness and equality. However, some variables whose behavior was inconsistent with the one expected, and others with opposite behaviors in the two areas were also found. Therefore, from the research findings, based on the data used, the methodological procedures and the statistical models adopted, it has been evidenced that value-added models improve the analysis reliability in comparison with models that use cross-sectional data, but they are still impracticable as tools for education system management, particularly for meritocratic use of their results. Hence, this research has corroborated the findings of other studies carried out over the world and has enabled us to state that the quality of the statistical modeling depends on the quality of data that it attempts to model, and it may generate distortions, establish unexpected relationships or lead to misleading conclusions. On the other hand, these resources may be used in the education system by providing important data for guiding public policies in a educative evaluation perspective, aiming at improving the quality of teaching offered by schools, teachers and other professionals that work in the school setting / Doutorado / Ensino e Práticas Culturais / Doutor em Educação

No description available.

Data from smartphones and wearable devices provide rich longitudinal information on participants and allow for causal inference for daily exposures and outcomes. However, informative missingness, latent variables, unmeasured confounding, and uneven data collection rates are common in mobile health studies and may introduce bias. In addition, there are likely violations of stationarity, a key assumption for traditional longitudinal methods. To overcome these challenges, we first propose an expectation maximization algorithm to adapt the conventional test for unit root non-stationarity to a context with missing data, and develop a sensitivity analysis for data missing not at random. Using our method, we identify a patient with bipolar spectrum disorder who has a unit root in their daily negative mood score data. We hypothesize the non-stationarity may result from the underlying latent disease states such as mania or depression, and thus we additionally develop a model to identify and control for latent modification and confounding. Specifically, we propose a hidden Markov model for individual causal inference which handles missing data in the outcome through marginalization and multiple imputation. We compare the performance of our proposed model with a frequentist and a Bayesian implementation to a naive approach in a simulation and application to a multi-year smartphone study of bipolar patients. We employ the approach to evaluate the individual effect of digital social activity on sleep duration across different latent disease states. Lastly, we employ functional data analysis methods to summarize overnight wrist actigraph data, to evaluate the role of sleep as a mediator between stress and positive mood. We demonstrate that functional principal component analysis identifies key information about sleep that is otherwise lost using a scalar representation of sleep duration.

Chronic obstructive pulmonary disease (COPD) and emphysema together are one of the leading causes of death in the United States and worldwide; meanwhile, breast cancer has the highest incidence and second-highest mortality burden of all cancers in women. Imaging markers relevant to each of these conditions are readily identifiable on chest computed tomography (CT): (1) visually-appreciable variants in airway tree structure exist which are associated with increased odds for development of COPD; (2) CT emphysema subtypes (CTES), based on lung texture and spatial features, have been identified by unsupervised clustering and correlate with functional measures and clinical outcomes; (3) dysanapsis, or the ratio of airway caliber to lung volume, is the strongest known predictor of COPD risk, and (4) breast density (i.e., the extent of fibroglandular tissue within the breast) is strongly associated with breast cancer risk. Machine- and deep-learning frameworks present an opportunity to address unmet needs in each of these directions, leveraging the data from large CT cohorts. Application of unsupervised learning approaches serves to discover new, image-based phenotypes. While topologic and geometric variation in the structure of the CT-resolved airway tree are well-described, tree- structural subtypes are not fully characterized. Similarly, while the clinical correlates of CTES have been described in large cohort studies, the association of CTES with structural and functional measures of the lung parenchyma are only partially described, and the time-dependent evolution of emphysematous lung texture has not been studied. Supervised approaches are required to automate CT image assessment, or to estimate CT- based measures from incomplete input data. While dysanapsis can be directly quantified on full- lung CT, the lungs are often only partially imaged in large CT datasets; total lung volume must then be regressed from the observed partial image. Breast density grades, meanwhile, are generally visually assessed, which is laborious to perform at scale. Moreover, current automated methods rely on segmentation followed by intensity thresholding, excluding higher-order features which may contribute to the radiologist assessment. In this thesis, we present a series of machine-learning methods which address each of these gaps in the field, using CT scans from the Multi-Ethnic Study of Atherosclerosis (MESA), the SubPopulations and InteRmediate Outcome Measures in COPD (SPIROMICS) Study, and an institutional chest CT dataset acquired at Columbia University Irving Medical Center. First, we design a novel graph-based clustering framework for identifying tree-structure subtypes in Billera-Holmes-Vogtmann (BHV) tree-space, using the airway trees segmented from the full-lung CT scans of MESA Lung Exam 5. We characterize the behavior of our clustering algorithm on a synthetic dataset, describe the geometric and topological variation across tree-structure clusters, and demonstrate the algorithm’s robustness to perturbation of the input dataset and graph tuning parameter. Second, in MESA Lung Exam 5 CT scans, we quantify the loss of small-diameter airway and pulmonary vessel branches within CTES-labeled lung tissue, demonstrating that depletion of these structures is concentrated within CTES regions, and that the magnitude of this effect is CTES-specific. In a sample of 278 SPIROMICS Visit 1 participants, we find that CTES demonstrate distinct patterns of gas trapping and functional small airways disease (fSAD) on expiratory CT imaging. In the CT scans of SPIROMICS participants imaged at Visit 1 and Visit 5, we update the CTES clustering pipeline to identify longitudinal emphysema patterns (LEPs), which refine CTES by defining subphenotypes informative of time-dependent texture change. Third, we develop a multi-view convolutional neural network (CNN) model to estimate total lung volume (TLV) from cardiac CT scans and lung masks in MESA Lung Exam 5. We demonstrate that our model outperforms regression on imaged lung volume, and is robust to same- day repeated imaging and longitudinal follow-up within MESA. Our model is directly applicable to multiple large-scale cohorts containing cardiac CT and totaling over ten thousand participants. Finally, we design a 3-D CNN model for end-to-end automated breast density assessment on chest CT, trained and evaluated on an institutional chest CT dataset of patients imaged at Columbia University Irving Medical Center. We incorporate ordinal regression frameworks for density grade prediction which outperform binary or multi-class classification objectives, and we demonstrate that model performance on identifying high breast density is comparable to the inter-rater reliability of expert radiologists on this task.

Financial strain and substance use are widespread in the United States (US), as many adults struggle to meet basic financial needs. There are many contributing causes of financial strain, such as unemployment or job loss and poverty or income loss. However, financial strain is distinct from its sources, as the experience of strain (i.e., inability to pay for bills, housing, healthcare, or food) may be necessary to produce a stress response that prompts substance use as a coping mechanism. Studies examining the relationship between financial strain and substance use have predominantly focused on the association between financial strain and alcohol use or acute financial strain due to brief changes in employment or income. However, there is a need to examine whether the relationship between financial strain and substance use varies by substance and the duration of financial strain (e.g., acute or brief financial strain, chronic or persistent financial strain, or intermittent financial strain due to repeated periods of financial strain). Several social safety net programs may mitigate financial strain, such as Unemployment Insurance, the Temporary Assistance for Needy Families program, the Supplemental Nutritional Assistance Program, and economic policies such as the federal Earned Income Tax Credit (EITC). Anti-poverty programs such as the federal EITC may provide an opportunity to reduce financial strain and subsequent substance use by providing financial support to low-income working populations. The purpose of this dissertation was to provide novel insights into the associations between acute, chronic, or intermittent financial strain and substance use and whether federal EITC eligibility altered these relationships. To achieve these goals, I had four aims. First, I sought to understand the associations between substance use and financial strain and its contributing causes (e.g., unemployment, poverty, etc.) by conducting a scoping review of the substance use literature (Aim 1; Chapter 2). Second, I examined the associations between acute, chronic, and intermittent financial strain and past month binge alcohol, cannabis, central nervous system (CNS) depressant (i.e., painkillers, sedatives, or tranquilizers), stimulant,and poly-drug use and whether these associations varied by sex (Aim 2.1; Chapter 3.1). Third, I examined the relationships between any financial strain and chronic financial strain on past month cannabis, CNS depressant, stimulant, and poly-drug use, adjusting for the time-dependent relationship between financial strain and drug use (Aim 2.2; Chapter 3.2). Fourth, I examined whether federal EITC eligibility was associated with short-term changes in financial strain, cannabis use, and CNS depressant use and whether this varied by state EITC policies or sex (Aim 3; Chapter 4). I conducted the scoping review presented in Chapter 2 in July-August 2023, searching for literature on the US, published since 2001 in PubMed, EBSCOhost [APA PsycINFO Database, EconLit with Full Text Database, SocINDEX with Full Text Database, Social Sciences Full Text Database], Web of Science, and Scopus. Two reviewers screened each abstract and title and conducted the full-text review. I extracted study characteristics and synthesized and evaluated evidence of the relationships between financial strain and its contributing causes and substance use. I found that more than one-third of studies examined employment-related measures (i.e., unemployment, job loss, or duration of unemployment) as indicators of financial strain, approximately 31% examined income-related measures (e.g., poverty, income loss) as indicators of financial strain, and only one-fifth of studies directly assessed financial strain, with more than half of those studies using a single-item question to assess financial strain. Job loss and duration of unemployment were generally associated with increased tobacco, binge alcohol, cannabis, opioid, drug, and substance use. Income-related indicators of financial strain were positively associated with tobacco, binge alcohol, and opioid use but inversely associated with any alcohol and cannabis use. Most studies found that financial strain was positively associated with tobacco and binge alcohol use. There was also evidence of a bi-directional relationship between alcohol and drug use and disorder with unemployment. My empirical aims leveraged individual- and state-level data from 5 waves (2013-2019) of the Population Assessment of Tobacco and Health study, a longitudinal cohort of >30,000 US adults. In Chapter 3.1, I characterized financial strain exposure patterns (i.e., none, acute, chronic, and intermittent) across Waves 1-5 and used survey-weighted multinomial logistic regression models to estimate the adjusted relative risk ratio of past month binge alcohol, cannabis, CNS depressants, stimulants, and poly-drug use at Wave 5 by financial strain exposure patterns, and whether this varied by sex. Consistent with prior research, I found that the relationship between financial strain and substance use varied by substance. Acute and intermittent financial strain were associated with an increased likelihood of binge alcohol, stimulant, and poly-drug use, whereas chronic financial strain was associated with an increased likelihood of cannabis or CNS depressant use. I found modest or no sex differences in the relationship between financial strain duration and binge alcohol, cannabis, CNS depressant, stimulant, or poly-drug use. Findings were robust in sensitivity analyses when I varied the number of waves considered chronic financial strain. Together, findings suggest that individuals experiencing a period of financial strain may be at greater risk of past month binge alcohol or drug use, and the risk for cannabis and CNS depressant use may increase with financial strain duration. In Chapter 3.2, I used longitudinal targeted maximum likelihood estimation methods to account for time- dependent confounding between financial strain and drug use and to estimate the association between any financial strain or chronic financial strain and cannabis, CNS depressants, stimulants, and poly-drug use across Waves 1-5. This doubly robust method allowed me to predict the average expected outcomes if everyone had never experienced financial strain across the study period (i.e., a never financial strain scenario) and if everyone had experienced financial strain at every time point across the study period (i.e., a chronic financial strain scenario). I then compared the expected outcomes under the never financial strain scenario to outcomes in the observed data to estimate the association between any financial strain and drug use. I found that if everyone had never experienced financial strain, the prevalence of past month cannabis use would be lower than the prevalence of past month cannabis use observed in the data. When I compared the expected outcomes under the never financial strain scenario to the expected outcomes in the chronic financial strain scenario, I found that the likelihood of past month cannabis or CNS depressant use would be significantly higher if everyone had experienced chronic financial strain across all five waves of data collection. These findings confirm the results in Chapter 3.1, suggesting that the association between chronic financial strain and drug use remained elevated after accounting for the potentially reinforcing relationship between financial strain and drug use over time. These analyses additionally revealed that the risk of cannabis use would have been slightly lower if no one had ever experienced financial strain compared with any financial strain. In Chapter 4, I examined changes in financial strain, cannabis use, and CNS depressant use associated with federal EITC eligibility during the EITC disbursement period. EITC could be considered a short-term intervention for financial strain and could provide temporary relief to people experiencing chronic financial strain, helping them transition from chronic to intermittent financial strain and lowering the likelihood of cannabis and CNS depressant use. I used survey participants’ interview dates to assess if EITC-eligible people who were randomly interviewed during the EITC disbursement period (February-April) had a lower risk of financial strain compared with EITC-eligible people interviewed outside the disbursement period (May-January) and EITC-ineligible people, using linear binomial models with a two-way interaction term for EITC eligibility and the EITC disbursement period. I then assessed whether changes in financial strain, cannabis use, and CNS depressant use associated with EITC eligibility during the EITC disbursement period varied by state refundable EITC policies. To do this, I used linear binomial models with a three-way interaction term between EITC eligibility, EITC disbursement period, and refundable state EITC policies. Finally, I conducted stratified analyses by sex to examine whether changes in financial strain, cannabis use, and CNS depressant use associated with EITC eligibility during the EITC disbursement period varied by men and women. Results indicated that receiving an EITC refund of at least $500 or more was associated with decreased financial strain, particularly among women. However, EITC eligibility during the EITC disbursement period was not significantly associated with past month cannabis or CNS depressant (i.e., painkiller, sedative, or tranquilizer) use overall or by sex at Wave 1. Changes in financial strain, cannabis use, and CNS depressant use associated with federal EITC eligibility during the disbursement period did not vary by whether the participant’s state of residence offered an additional refundable EITC. Findings from this dissertation provide empirical support for financial strain as a potential predictor of binge alcohol, cannabis, CNS depressant, stimulant, and poly-drug use. This relationship varied by duration of strain, and the association between chronic financial strain and drug use remained elevated when adjusting for potential time-varying confounding in this relationship. I also found that the likelihood of cannabis use would decrease if no one had ever experienced financial strain. When I examined the federal EITC as a potential short-term intervention for financial strain, I found that refunds of at least $500 or more were associated with decreased financial strain without increasing cannabis or CNS depressant use in the overall population. Thus, expanded and more generous income support policies such as the EITC may be effective tools to intervene on financial strain.

Le turnover au sein des organisations, à un faible niveau, peut être bénéfique pour le renouvellement des ressources. Mais en atteignant des sommets, il perturbe l’organisation et oblige à une gestion de l’immédiat, nuisible pour la performance des hommes et de l’entreprise. Le groupe Veolia Transport souffre de cet aléa organisationnel de manière récurrente, quand le cœur de son activité est basé sur la délégation de service public, où la qualité des hommes est gage de la performance de l’organisation. En appui de cet impératif qualitatif, le groupe promeut une politique générale de formation ambitieuse afin d’intégrer, professionnaliser et fidéliser ses collaborateurs. Le fer de lance de cette politique est l’alternance, la formation diplômante en alternance. Dès lors, l’alternance impacte-t-elle à la baisse le turnover ? À travers une démarche proche d’un design quasi-expérimental, mobilisant une enquête dite longitudinale par questionnaire et l’appui d’un groupe de contrôle je teste, sur les exploitants du groupe en France, un design théorique explicatif de la relation alternance-turnover, par le prisme de l’implication organisationnelle. En proposant une définition gestionnaire de l’alternance, mes résultats démontrent le développement d’une relation dialectique individu-organisation par l’alternance, conditionnée par la mise en exergue d’un contrat psychologique fort et d’un double tutorat organisationnel. En croisant mes résultats avec 18 entretiens dits de validation, il ressort que l’alternance développe une socialisation organisationnelle partielle, en raison de la difficulté à comprendre pleinement son rôle en fin de cursus. Un glissement de la fonction tutorale en un système tutoral apparait, où la communauté de travail aide à l’apprentissage du métier, quand le supérieur-tuteur conserve un rôle de mentor. Enfin, l’alternance possède un effet positif sur l’intention de quitter, en créant les conditions du développement de l’implication organisationnelle, à la condition d’une gestion dédiée. Dans ce dessein, je propose la mise en place d’une gestion spécifique des alternants, en ne considérant plus l’alternance comme un outil de formation mais comme un outil de gestion des ressources humaines, créateur de potentiels. / Within the organizations, the turnover when limited can have a positive influence on the resources’ renewal. But when it grows, it badly affects the organization leading to a management of the immediate, jeopardizing people and company’s effectiveness. Veolia Group is suffering from this regular organizational move, when its core business is made of services and where people are the key quality asset. On top of that, the company encourages an ambitious training policy in order to integrate, professionalize, and retain employees. To be successful, the company relies on the classroom learning (alternation). Therefore, can we claim that alternation has a direct impact on turnover’s decrease?Thanks to a quasi-experimental design approach, using an investigation so called longitudinal per questionnaire strengthened by a group of control, a theoretical design explaining the relationship alternation-turnover in the organizational frame was tested in France over a French population of production units. Alternation shows the emergence of a dialectical relationship human being-organization, monitored by the creation of a strong psychological contract and a double tutorial system. Mixing my results with 18 interviews of so called validation, it appears that alternation develops an organizational socialization only partial, due to the difficulty in the understanding of its own role at the end of the journey. The shift from a tutorial function to a tutorial system is also highlighted in my results, where the learning of the job is supported by the working community and the lead-tutor continues playing a mentor role. Lastly, according my study, when well managed, alternation can prevent the turnover, by supporting the development of a strong involvement within the organization. Therefore, I suggest the set up of a specific management unit for alternates, considering not anymore the alternation as a training tool but also as human resource tool enabling talents’ discovery.

This dissertation focuses on the formation, growth and discontinuance of business populations and firms in Sweden during the 20th century. It addresses some key issues in the domain of economic and social sciences, and in particular entrepreneurship and small business research: if and when firms grow, stagnate and decline, as well as how long firms survive and when they are likely to disband. Previous research has primarily analyzed these questions from a short time frame. Further, an individual or firm-oriented focus is commonly assumed. In that, alternative or complementary explanations to the growth and survival of firms may be disregarded. In contrast to much previous research, this dissertation assumes a micro-to-macro, longitudinal and demographic population approach. The period of investigation is over one hundred years. In addressing the growth and survival of firms, it takes into account the impact of firm-specific structural factors (such as firm age and size), generation (cohort) effects, as well as the influence of macroeconomic, exogenous factors. Further, the relationship between managerial/ownership succession and firm performance is also addressed. Both cross-sectional and longitudinal databases are employed in the dissertation. Its main empirical material consists of unique longitudinal data on new business firms, traced at the firm level from their birth to their termination. More specifically, seven birth cohorts – generations – of approximately 2,200 firms founded in 1899, 1909, 1912, 1921, 1930, 1942 and 1950 are included. The main findings show that ownership/management succession in firms had a quite weak correlation with firm performance and survival. At least at an aggregate level, and with some exceptions, it is debatable if the loss and replacement of owner-managers in small and in larger firms have any observable effects on firm performance. Furthermore, macroeconomic phenomena influence the conditions of individual firms as well as populations/aggregates of businesses. Both the growth and termination of firms and firm populations are found to be related to real economic (environmental) conditions; e.g. favorable macroeconomic conditions implied that firms grew in size. At the same time, under certain circumstances, the influence of structural variables (firm age and size) – as suggested in much previous research – is found to be of importance. As concerns firm growth, as well as firm termination, the economic environment and structural factors interact. These findings challenges individual or firm-level research that mainly focus on personal traits and behaviors in explaining firm success and failure. Other previous assumptions are also challenged when taking a longer time perspective into consideration. For decades, organization and business research have acknowledged a liability of newness and of size for business firms. While this might be true under some conditions, this liability of newness is falsified in the study: the termination behavior of some firm generations did not correspond with these assumptions. Thus, the perspectives and methodology applied in the dissertation complement earlier approaches in entrepreneurship and small business research.

Indiana University-Purdue University Indianapolis (IUPUI) / Symptoms of post-traumatic stress disorder (PTSD) are a common comorbidity in veterans seeking treatment of chronic musculoskeletal pain (CMP). However, little is known regarding the mutual influence of PTSD and CMP in this population. Using cross-sectional and longitudinal data from a randomized clinical trial evaluating a stepped care intervention for CMP in Iraq/Afghanistan veterans (ESCAPE), this dissertation examined the relationships between PTSD and CMP along with other factors including depression, anxiety, catastrophizing and health-related quality of life. The Classification and Regression Tree (CART) analysis was conducted to identify key factors associated with baseline PTSD besides CMP severity. A series of statistical analyses including logistical regression analysis, mixed model repeated measure analysis, confirmatory factor analysis and cross-lagged panel analysis via structural equation modeling were conducted to test five competing models of PTSD symptom clusters, and to examine the mutual influences of PTSD symptom clusters and CMP outcomes. Results showed baseline pain intensity and pain disability predicted PTSD at 9 months. And baseline PTSD predicted improvement of pain disability at 9 months. Moreover, direct relationships were found between PTSD and the disability component of CMP, and indirect relationships were found between PTSD, CMP and CMP components (intensity and disability) mediated by depression, anxiety and pain catastrophizing. Finally, the coexistence of PTSD and more severe pain was associated with worse SF-36 Physical Component Summary (PCS) and Mental Component Summary (MCS) scores. Together these findings provided empirical support for the mutual maintenance theory.

Indiana University-Purdue University Indianapolis (IUPUI) / Epidemiologic and clinical studies routinely collect longitudinal measures of multiple outcomes. These longitudinal outcomes can be used to establish the temporal order of relevant biological processes and their association with the onset of clinical symptoms. In the first part of this thesis, we proposed to use bivariate change point models for two longitudinal outcomes with a focus on estimating the correlation between the two change points. We adopted a Bayesian approach for parameter estimation and inference. In the second part, we considered the situation when time-to-event outcome is also collected along with multiple longitudinal biomarkers measured until the occurrence of the event or censoring. Joint models for longitudinal and time-to-event data can be used to estimate the association between the characteristics of the longitudinal measures over time and survival time. We developed a maximum-likelihood method to joint model multiple longitudinal biomarkers and a time-to-event outcome. In addition, we focused on predicting conditional survival probabilities and evaluating the predictive accuracy of multiple longitudinal biomarkers in the joint modeling framework. We assessed the performance of the proposed methods in simulation studies and applied the new methods to data sets from two cohort studies. / National Institutes of Health (NIH) Grants R01 AG019181, R24 MH080827, P30 AG10133, R01 AG09956.

Indiana University-Purdue University Indianapolis (IUPUI) / Magnetic resonance (MR) spectroscopy is a neuroimaging technique. It is widely used to quantify the concentration of important metabolites in a brain tissue. Imbalance in concentration of brain metabolites has been found to be associated with development of neurological impairment. There has been increasing trend of using MR spectroscopy as a diagnosis tool for neurological disorders. We established statistical methodology to analyze data obtained from the MR spectroscopy in the context of the HIV associated neurological disorder. First, we have developed novel methodology to study the association of marker of neurological disorder with MR spectrum from brain and how this association evolves with time. The entire problem fits into the framework of scalar-on-function regression model with individual spectrum being the functional predictor. We have extended one of the existing cross-sectional scalar-on-function regression techniques to longitudinal set-up. Advantage of proposed method includes: 1) ability to model flexible time-varying association between response and functional predictor and (2) ability to incorporate prior information. Second part of research attempts to study the influence of the clinical and demographic factors on the progression of brain metabolites over time. In order to understand the influence of these factors in fully non-parametric way, we proposed LongCART algorithm to construct regression tree with longitudinal data. Such a regression tree helps to identify smaller subpopulations (characterized by baseline factors) with differential longitudinal profile and hence helps us to identify influence of baseline factors. Advantage of LongCART algorithm includes: (1) it maintains of type-I error in determining best split, (2) substantially reduces computation time and (2) applicable even observations are taken at subject-specific time-points. Finally, we carried out an in-depth analysis of longitudinal changes in the brain metabolite concentrations in three brain regions, namely, white matter, gray matter and basal ganglia in chronically infected HIV patients enrolled in HIV Neuroimaging Consortium study. We studied the influence of important baseline factors (clinical and demographic) on these longitudinal profiles of brain metabolites using LongCART algorithm in order to identify subgroup of patients at higher risk of neurological impairment. / Partial research support was provided by the National Institutes of Health grants U01-MH083545, R01-CA126205 and U01-CA086368