published_or_final_version / Statistics and Actuarial Science / Doctoral / Doctor of Philosophy

In this report, I study familial longitudinal count data with a Poisson regression model. The data is collected from individuals who are nested in families. I focus on two main issues to fit a model. The first one is the large number of excess zeros and the second one is multi-level random effects. My approach for solving these problems are to use either Zero Inflated Poisson (ZIP) or Negative Binomial (NB) models to control for the excess zeros which allow for estimation of another parameter for over dispersion while developing the model with individual and familial random effects. First, I use a Poisson regression model with only main effects. After that, I fit a ZIP model to control for the extra zeros. I provide information about general form of the exponential families and a discussion about the dispersion parameter. I also fit a Negative Binomial model instead of the ZIP model. I also build these models with only individual random effects and with both individual and familial random effects as well. I discuss the generalized estimating equation (GEE) approach to estimate the parameters of a generalized linear model with auto regressive correlation between outcomes. / text

No description available.

No description available.

A common goal in medical research is to determine the effect that a treatment has on subjects over time. Unfortunately, the analysis of data from such clinical trials often omits several aspects of the study design, leading to incorrect or misleading conclusions. In this paper, a major objective is to show via case studies that randomized controlled trials with longitudinal designs must account for correlation and clustering among observations in order to make proper statistical inference. Further, the effects of outliers in a multi-center, randomized controlled trial with multiple layers of clustering are examined and strategies for detecting and dealing with outlying observations and clusters are discussed. / McAnulty College and Graduate School of Liberal Arts; / Computational Mathematics / MS; / Thesis;

Chiu Chu-hing, Catherine. / Thesis (M.Ph.)--Chinese University of Hong Kong, 1985 / Bibliography: leaves 90-97

Family therapists have proposed that specific types of family interactions are dysfunctional for the family system and can produce long-term negative effects for the child. They further propose that, for healthy family functioning to be maintained, parental alliances must be sustained and excessive cross-generational coalitions (parent-child alliances) must be blocked. This fundamental assumption, proposed by family therapists, has rarely been empirically tested. The purpose of this study was to investigate the relationship between family interactional patterns and the misconduct of adolescent males. Misconduct by the adolescent was defined by the youth having contact with the police due to delinquent behavior. Specifically, this study was concerned with determining if cross-generational coalitions, witnessed in parent-child interactions, were predictive of later antisocial behavior on the part of the adolescent while controlling for family structure, family problem solving, marital satisfaction, parental conflict, and child externality. Subjects were 68 families consisting of mother, father or stepfather, and son. The first family interaction assessment took place in 1984-1985 when the child was 9.7 years old. The second assessment of family interaction occurred two years later. Follow-up data on the adolescent's delinquent behavior, assessed through county court records, was last collected in 1991 when subjects were 15 to 16 years old. Families were paid for their participation as part of their involvement in a larger study (Capaldi & Patterson, 1987). The results of logistic and multiple regression analyses indicated no association between parent-child coalitions and occurrence or the severity of delinquent behavior. Both analyses did, however, find that family problem-solving skills and a non-intact family structure were significant predictors of later delinquency and of the severity of the delinquency by the adolescent. / Graduation date: 1993

Many studies have attempted to link antimicrobial use in food animal agriculture with an increased risk of antimicrobial-resistant (AR) bacterial levels in humans. Our data arise from longitudinal aggregated fecal samples in a 3-year cohort study of vertically integrated populations of human workers and consumers, and swine. Human and swine E. coli isolates (N = 2130 and 3485, respectively) were tested for antimicrobial susceptibility using the SensititreTM broth microdilution system. The associations between AR prevalence for each antimicrobial agent, multi-drug resistant E. coli, or multivariate AR E. coli, and the risk factors (host species, production type (swine), vocation (human swine worker versus non-worker), and season) in the study were assessed using generalized estimating equations (GEE), GLM with multinomial distribution, or GEE in a multivariate model using a SAS® macro to adjust for the correlated AR phenotypes. There were significant (p < 0.05) differences in AR isolates: 1) between host-species with swine at higher risk for ceftiofur, chloramphenicol, gentamicin, kanamycin, streptomycin, sulfisoxazole, and tetracycline. The prevalence of ciprofloxacin, nalidixic acid, and trimethoprim/sulfamethoxazole resistance were higher among human isolates, 2) swine production group was significantly associated with AR with purchased boars, nursery piglets, and breeding boars at a higher risk of resistance to streptomycin and tetracycline, and 3) human swine worker cohorts exhibited an elevated tetracycline prevalence, but lowered sulfisoxazole prevalence when compared to nonworkers. High variability among seasonal samples over the 3-year period was observed. There were significant differences in multiple resistance isolates between host species, with swine at higher risk than humans of carrying multi-resistant strains; however, no significant differences in multiple resistance isolates within humans by vocation or within swine by production group. The odds-ratios, adjusted for multivariate dependence of individual AR phenotypes, were increased relative to unadjusted oddsratios among 1) swine as compared to human for tetracycline (OR = 21.8 vs. 19.6), and 2) increased significantly among swine-workers as compared to non-workers only for tetracycline (OR = 1.4 vs. 1.3). Occupational exposure to swine-rearing facilities appears to be associated with an increased relative odds for the prevalence of tetracycline resistance compared to non-workers.

Thesis (Ph. D.)--Oregon State University, 2010. / Printout. Includes bibliographical references (leaves 45-50). Also available on the World Wide Web.

There is emerging evidence of the significance of paternal mental health problems as well as its influence on spouses and children. However, current research attention mainly focuses on paternal postpartum depression. The present study aims at determining the prevalence of paternal mental health problems, identifying risk factors and the relationship among various risk factors across different stages of pregnancy and the postpartum period. In particular, it attempts to examine the significance of gender role and traditionalism-modernity as personality traits for paternal mental health problems in the perinatal period as well as fill the current gap in knowledge on the impact of paternal mental health problems on paternal-fetal attachment. Given the relative lack of data in the existing literature, the present study involved two phases of investigation combining qualitative and quantitative approaches to provide a more systematic understanding on the nature of paternal mental health problems. In phase 1, a qualitative study was conducted on 31 expectant fathers as the hypothesis-generating ground work. Subsequently, a total of 622 expectant couples were recruited and followed up in a large-scale survey in Phase 2. Expectant couples were recruited from regional antenatal clinics in Tsan Yuk Hospital and Queen Mary Hospital. In the quantitative component, the expectant couples were assessed using standardized and validated psychological instruments on 3 time points including early pregnancy, late pregnancy and six week postpartum. The results showed that a significant proportion of expectant fathers manifested anxiety, and depressive symptoms, high level of perceived stress, psychosomatic symptoms, fatigue and tobacco consumption during the perinatal period. Different demographic risk factors for the outcome variables varied across different time points. Psychosocial risk factors were consistently associated with outcome variables in different time points. Maternal depression and anxiety were not only predictors of paternal psychological distress but also the predictor of higher level of paternal psychosomatic symptoms and fatigue severity. Gender role and Traditionalism-modernity were found to be significant moderators for paternal mental health problems. Undifferentiated, masculine fathers and fathers who adhere to traditional values were at higher risk of mental health problems during the transition of fatherhood. A mixed effect model analysis showed that paternal anxiety, depression and psychosomatic symptoms changes over time and was higher during the antenatal period. The relationships between paternal and maternal anxiety and depressive symptoms were significant across different time points even after controlling for confounding factors and the correlations were found to be strongest at six weeks postpartum. Findings also showed that paternal mental health problems had a detrimental effect on paternal-fetal bonding from antenatal to postpartum period. The present study points to the need for greater research and clinical attention to paternal mental health problems given that paternal mental health problems is a highly prevalent problem with changing course and is related to maternal well-being and paternal-fetal attachment. The present findings contributes to theoretical basis of the relationships of personality traits, risk factors and mental health problems and have implications for the design of effective identification, prevention, and interventions of these clinical problems. / published_or_final_version / Psychiatry / Doctoral / Doctor of Philosophy