Clinical practice should be based on the best available evidence. Ideally such evidence is obtained through rigorously conducted, purpose-designed clinical studies such as randomised controlled trials and prospective cohort studies. However gathering information in this way requires a massive effort, can be prohibitively expensive, is time consuming, and may not always be ethical or practicable. When answers are needed urgently and purpose-designed prospective studies are not feasible, retrospective healthcare data may offer the best evidence there is. But can we rely on analysis with such data to give us meaningful answers? The current thesis studies this question through analysis with repeated psychological symptom screening data that were routinely collected from over 20,000 outpatients who attended selected oncology clinics in Scotland. Linked to patients’ oncology records these data offer a unique opportunity to study the progress of distress symptoms on an unprecedented scale in this population. However, the limitations to such routinely collected observational healthcare data are many. We approach the analysis within a missing data context and develop a Bayesian model in WinBUGS to estimate the posterior predictive distribution for the incomplete longitudinal response and covariate data under both Missing At Random and Missing Not At Random mechanisms and use this model to generate multiply imputed datasets for further frequentist analysis. Additional to the routinely collected screening data we also present a purpose-designed, prospective cohort study of distress symptoms in the same cancer outpatient population. This study collected distress outcome scores from enrolled patients at regular intervals and with very little missing data. Consequently it contained many of the features that were lacking in the routinely collected screening data and provided a useful contrast, offering an insight into how the screening data might have been were it not for the limitations. We evaluate the extent to which it was possible to reproduce the clinical study results with the analysis of the observational screening data. Lastly, using the modelling strategy previously developed we analyse the abundant screening data to estimate the prevalence of depression in a cancer outpatient population and the associations with demographic and clinical characteristics, thereby addressing important clinical research questions that have not been adequately studied elsewhere. The thesis concludes that analysis with observational healthcare data can potentially be advanced considerably with the use of flexible and innovative modelling techniques now made practicable with modern computing power.

In this submission, five articles are presented examining one theme: diagnosis of autism spectrum disorders (ASD) in children. Three articles provides perspectives on various social and medical factors that influence the diagnosis of ASD, and the others examine social and behavioural outcomes for children diagnosed with ASD. One article provides an in depth examination of the dilemmas of diagnosis from a parental perspective. The research utilized both qualitative and quantitative methods. A secondary analysis of a longitudinal birth cohort study revealed that there were a number of children who had autistic traits equally severe as those with clinical diagnosis. Further analysis exposed a possible gender bias in diagnosis. Outcomes for children with ASD diagnoses were worse than for those without diagnoses but with comparable behaviours as preschoolers. ASD diagnosis apparently had no positive effect on the developmental trajectory of prosocial behaviour. The implications of these results are discussed. Analysis of qualitative data collected in semi-structured interviews with parents of both diagnosed and undiagnosed children exposed dilemmas faced by parents as they contemplated an ASD diagnosis and highlighted parental action to de-stigmatise the condition after diagnosis had been applied. The body of work as a whole falls at the junction of clinical and educational psychology, developmental psychology, social psychology, social psychiatry, sociology and epidemiology. It draws attention to a number of social processes that contribute to ASD diagnosis. Overall, it is argued, the work supports the conceptualisation of ASD as both a biologically and socially determined condition.

<p> In this report we explore how the effect of meteorological factors, allergens, and air pollution on respiratory conditions in children using longitudinal data. Our analysis makes use of a dataset from the DAVIS study in southern Ontario. The response variables are children's lower respiratory tract (URT) and upper respiratory tract (URT) scores. The explanatory variables are readings of various meteorological, allergen, and air pollution factors. First we make use of generalized estimating equations to find the main factors that are associated with certain respiratory conditions in children as measured by LRT and URT scores. Then we determine whether there are any interactions between the significant factors associated with LRT /URT scores. Comparisons between case and control groups are made to determine whether children with asthma are more sensitive to any of the changes in meteorological, allergen, and air pollution factors. The analysis results show that the significant factor that is associated with LRT scores for children with asthma is the two-day lag daily average changes in air pressure. On average an increase in air pressure will result in an increase in children's LRT scores. The interaction terms that remained in the final model show some degree of significance but without strong evidence to support it. Children in the case groups are more sensitive to meteorological factors, allergens, and air pollution than the children in control groups. </p> / Thesis / Master of Science (MSc)

Dementia is increasing recognized as a major problem to public health worldwide. Prevention and treatment strategies are in critical need. Nowadays, research for dementia usually featured as complex longitudinal studies, which provide extensive information and also propose challenge to statistical methodology. The purpose of this dissertation research was to apply statistical methodology in the field of dementia to strengthen the understanding of dementia from three perspectives: 1) Application of statistical methodology to investigate the association between potential risk factors and incident dementia. 2) Application of statistical methodology to analyze changes over time, or trajectory, in cognitive tests and symptoms. 3) Application of statistical learning methods to predict development of dementia in the future. Prevention of Alzheimer’s disease with Vitamin E and Selenium (PREADViSE) (7547 subjects included) and Alzheimer’s disease Neuroimaging Initiative (ADNI) (591 participants included) were used in this dissertation. The first study, “Self-reported sleep apnea and dementia risk: Findings from the PREADViSE Alzheimer’s disease prevention trial ”, shows that self-reported baseline history of sleep apnea was borderline significantly associated with risk of dementia after adjustment for confounding. Stratified analysis by APOE ε4 carrier status showed that baseline history of sleep apnea was associated with significantly increased risk of dementia in APOE ε4 non-carriers. The second study, “comparison of trajectories of episodic memory for over 10 years between baseline normal and MCI ADNI subjects,” shows that estimated 30% normal subjects at baseline assigned to group 3 and 6 stay stable for over 9 years, and normal subjects at baseline assigned to Group 1 (18.18%) and Group 5 (16.67%) were more likely to develop into dementia. In contrast to groups identified for normal subjects, all trajectory groups for MCI subjects at baseline showed the tendency to decline. The third study, “comparison between neural network and logistic regression in PREADViSE trial,” demonstrates that neural network has slightly better predictive performance than logistic regression, and also it can reveal complex relationships among covariates. In third study, the effect of years of education on response variable depends on years of age, status of APOE ɛ4 allele and memory change.

Ankylosing spondylitis (AS) is a chronic, inflammatory rheumatic disease that has a highly variable disease course. Three biologic agents, adalimumab, etanercept, and infliximab, have been developed for the treatment of AS. We conducted three studies: 1) an exploratory analysis of a year-long longitudinal dataset to gain insight into the variability of disease activity, physical function, and well-being and to explore the relationship between these outcome measures; 2) a systematic review of the available evidence for the efficacy of biologic treatment; 3) a systematic review of potential adverse effects of this treatment. We found that repeated measures of disease activity, function and well-being fluctuate considerably between patients, with complex patterns occurring over time within patients. There was mostly high quality evidence that these biologics are efficacious against placebo. We did not find evidence of an increase in serious adverse events or serious infections from short-term randomized controlled trials.

Title: The influence of swimming part of triathlon the total performance in competitions of world championships 2000-2010. Objectives: The aim is to investigate the influence of the swimming part of the total performance in triathlon in the lost 10 years. Methods: The relation between two variables (times from swimming and the total time of competitions from competitions of world championships 2000-2010 were described with the help of linear regression. Individual years were then compared in a point graph. Results: Our hypothese were confirmmed. It means that the result from the swimming part explains the result from the competition in less than 30%, which means the R2 is smaller than 0,3. We also found, that the influnence of the swimming part on the final result in triathlon has not changed in the last years. The relationship times in the swimming part and total (final) times was closest in 2005 (R2= 0,21). In other years, this dependence is lower, R2 was smaller than 0,07. Keywords: linear regression, longitudinal analysis, performance

Thesis advisor: Sara M. Moorman / There is much prior research on the benefits of marriage for adults, including for mental and physical health (Carr and Springer 2010). Further research has demonstrated that the quality of one’s marriage provides benefits, and not merely the status itself (see Carr and Springer 2010; Proulx, Helms, and Buehler 2007). A close, salient relationship such as marriage is not experienced in isolation, but is rather an interpersonal system, where the characteristics, feelings, and opinions of each partner can influence the other (Berscheid and Ammazzalorso 2001; Carr et al. 2014; Moorman 2016). However, less research has been performed that takes advantage of dyadic data to determine whether and how a partner’s marital quality may affect one’s own well-being (Carr et al. 2014; Kenny 1996). Moreover, emotional experiences rarely remain truly private; individuals unconsciously signal and express their feelings to others, and can even transmit these emotional experiences to close social partners (Christakis and Fowler 2013; Hatfield, Cacioppo, and Rapson 1994). The present dissertation examines the associations among older husbands’ and wives’ marital quality and well-being, using two sources of dyadic data, a range of measures of marital quality and well-being, and advanced analytic strategies appropriate for longitudinal and cross-sectional data. Older couples can differ from their younger and midlife counterparts, as both men and women trim their broader social networks in later life and increasingly focus on their closest and most rewarding relationships, such as marriage (Carstensen, Isaacowitz, and Charles 1999; Mancini and Bonanno 2006). Gendered roles may shift in later life, as well, as older adults cease activities such as child-rearing and full-time employment (Bookwala 2012). Thus, potential differences according to gender are also explicitly tested. The results of this dissertation will shed greater light on how older couples’ perceptions of marital quality influence various aspects of spouses’ well-being, cross-sectionally and over time. Mutual Influence and Older Married Adults’ Anxiety Symptoms: Results from The Irish Longitudinal Study on Ageing analyzes cross-sectional dyadic data from 1,114 married older couples surveyed in the initial wave of The Irish Longitudinal Study on Ageing (TILDA; Kenny 2014), 2009-2011. Dyadic structural equation models (SEM) examined the direct and indirect associations between husbands’ and wives’ reports of marital strain and generalized anxiety symptoms in later life. Findings revealed that perceptions of marital strain were related with husbands’ and wives’ own generalized anxiety symptoms. Further, husbands’ anxiety symptoms were significantly related with wives’ anxiety symptoms, and vice versa, illustrating bi-directional feedback. Lastly, husbands’ and wives’ perceptions of marital strain were significantly indirectly related with their partners’ anxiety symptoms, with these associations being mediated by spouses’ own anxiety symptoms. These results suggest that emotional contagion may be the pathway for partner effects of marital strain on spouses’ well-being. Findings also suggest that efforts to reduce anxiety symptoms may be most effective when taking marital context and quality into account. Two-Wave Dyadic Analysis of Marital Quality and Loneliness in Later Life: Results From The Irish Longitudinal Study on Ageing analyzes dyadic reports of marital quality and loneliness over a two-year period, using longitudinal dyadic data collected from 932 older married couples who participated in both of the first two waves of The Irish Longitudinal Study on Ageing (TILDA), collected from 2009-2013. Two-wave lagged dependent variable (LDV) models tested the cognitive perspective on loneliness, emotional contagion theory, and actor-partner interdependence by examining whether husbands’ and wives’ reports of marital quality and loneliness at baseline predicted both spouses’ loneliness two years later. Results indicated that one’s own perceptions of negative marital quality at baseline were related with greater loneliness after two years, supporting the cognitive perspective on loneliness. Further, both spouses’ reports of loneliness at baseline were related with loneliness two years later, supporting emotional contagion theory. Partners’ reports of marital quality were not related with future loneliness, failing to support actor-partner interdependence. Do “His” and “Her” Marriage Influence One Another? Older Spouses’ Marital Quality Over Four Years uses two-wave longitudinal data from the Disability and Use of Time (DUST) supplement to the Panel Study of Income Dynamics (PSID) to examine associations between husbands’ and wives’ reports of marital quality over a four-year period. The sample consisted of 209 older married couples who participated in both the 2009 and 2013 waves of DUST. Lagged dependent variable (LDV) models tested whether older husbands’ and wives’ perceptions of marital quality are themselves subject to emotional contagion, by examining whether baseline reports of marital quality were related with one’s own and a partner’s marital quality after four years. Results indicated that (a) husbands reported better marital quality than their wives in both 2009 and 2013, (b) for both husbands and wives, baseline marital quality was significantly related with both one’s own and one’s partner’s marital quality four years later, and (c) there were no differences in effects according to gender. These findings offer support for the framework of “his” and “her” marriage, as well as emotional contagion theory. Together, these papers examine whether and how older spouses’ reports of marital quality and well-being are associated with one another, with a particular emphasis on assessing emotional contagion as a potential explanation and mechanism for dyadic partner effects. The results of these articles contribute empirically and theoretically to the literature(s) on marital quality and well-being; spousal interdependence; and emotional contagion. I discuss the implications of these articles for theory and future research concerning marriage and well-being in later life.

Ankylosing spondylitis (AS) is a chronic, inflammatory rheumatic disease that has a highly variable disease course. Three biologic agents, adalimumab, etanercept, and infliximab, have been developed for the treatment of AS. We conducted three studies: 1) an exploratory analysis of a year-long longitudinal dataset to gain insight into the variability of disease activity, physical function, and well-being and to explore the relationship between these outcome measures; 2) a systematic review of the available evidence for the efficacy of biologic treatment; 3) a systematic review of potential adverse effects of this treatment. We found that repeated measures of disease activity, function and well-being fluctuate considerably between patients, with complex patterns occurring over time within patients. There was mostly high quality evidence that these biologics are efficacious against placebo. We did not find evidence of an increase in serious adverse events or serious infections from short-term randomized controlled trials.

Studies over the past couple of decades have depicted loneliness as a significant concern to physical health, although its meaning for overall health outcomes is still unclear. The precise impact of loneliness on life expectancy and on specific disease processes remains unknown. With regression modeling techniques, this thesis uses data from the Alameda County Health and Ways of Living Study to characterize the impact of loneliness on self-rated health, mortality, and fatalities from specific diseases. A key hypothesis is that loneliness as a health problem hinges on its persistence over time. This hypothesis is also tested by examining the reliability of the loneliness measure across the full 34 years of the survey. A second test is to examine its interplay with marital status as a mutable social circumstance. Results show that loneliness is a risk factor for poor self-rated health, non-ischemic cardiovascular diseases, cerebrovascular diseases, infections, and overall mortality. Results also show that loneliness need not be a stable problem across the life span in order to pose health risks. The reliability of the loneliness measure fades across time and levels of loneliness also vary with changes in marital status. Loneliness did not clearly mediate the impact of marital status on self-rated health, mortality, or specific causes of death.

Studies over the past couple of decades have depicted loneliness as a significant concern to physical health, although its meaning for overall health outcomes is still unclear. The precise impact of loneliness on life expectancy and on specific disease processes remains unknown. With regression modeling techniques, this thesis uses data from the Alameda County Health and Ways of Living Study to characterize the impact of loneliness on self-rated health, mortality, and fatalities from specific diseases. A key hypothesis is that loneliness as a health problem hinges on its persistence over time. This hypothesis is also tested by examining the reliability of the loneliness measure across the full 34 years of the survey. A second test is to examine its interplay with marital status as a mutable social circumstance. Results show that loneliness is a risk factor for poor self-rated health, non-ischemic cardiovascular diseases, cerebrovascular diseases, infections, and overall mortality. Results also show that loneliness need not be a stable problem across the life span in order to pose health risks. The reliability of the loneliness measure fades across time and levels of loneliness also vary with changes in marital status. Loneliness did not clearly mediate the impact of marital status on self-rated health, mortality, or specific causes of death.