No description available.

No description available.

No description available.

Semiparametric modeling is a combination of the parametric and nonparametric models in which some functions follow a known form and some others follow an unknown form. In this dissertation we made contributions to semiparametric modeling for matched case-crossover data. In matched case-crossover studies, it is generally accepted that the covariates on which a case and associated controls are matched cannot exert a confounding effect on independent predictors included in the conditional logistic regression model. Any stratum effect is removed by the conditioning on the fixed number of sets of the case and controls in the stratum. However, some matching covariates such as time, and/or spatial location often play an important role as an effect modification. Failure to include them makes incorrect statistical estimation, prediction and inference. Hence in this dissertation, we propose several approaches that will allow the inclusion of time and spatial location as well as other effect modifications such as heterogeneous subpopulations among the data. To address modification due to time, three methods are developed: the first is a parametric approach, the second is a semiparametric penalized approach and the third is a semiparametric Bayesian approach. We demonstrate the advantage of the one stage semiparametric approaches using both a simulation study and an epidemiological example of a 1-4 bi-directional case-crossover study of childhood aseptic meningitis with drinking water turbidity. To address modifications due to time and spatial location, two methods are developed: the first one is a semiparametric spatial-temporal varying coefficient model for a small number of locations. The second method is a semiparametric spatial-temporal varying coefficient model, and is appropriate when the number of locations among the subjects is medium to large. We demonstrate the accuracy of these approaches by using simulation studies, and when appropriate, an epidemiological example of a 1-4 bi-directional case-crossover study. Finally, to explore further effect modifications by heterogeneous subpopulations among strata we propose a nonparametric Bayesian approach constructed with Dirichlet process priors, which clusters subpopulations and assesses heterogeneity. We demonstrate the accuracy of our approach using a simulation study, as well a an example of a 1-4 bi-directional case-crossover study. / Ph. D.

No

Idiopathic normal pressure hydrocephalus (INPH) is a dementia treatable by insertion of a cerebrospinal fluid shunt. It has been suggested that INPH has similar pathophysiological mechanisms as cerebrovascular disease, but the vascular risk factor (VRF) profile of INPH patients has not been assessed using a modern epidemiological approach. The cognitive symptoms of INPH resemble the symptoms of depression, but the prevalence of depression among INPH patients is unknown. In addition, few studies investigate the impact of shunting on the quality of life (QoL), and no study has investigated the impact of comorbidity on QoL in INPH patients. The objective of this dissertation was to present the VRF profile of INPH and to investigate the hypothesis that INPH may be a subgroup of vascular dementia. Additional objectives were to assess the prevalence of depression in INPH patients and to investigate the impact of shunting and comorbidities on QoL in INPH. In the first cohort, the prevalence of possible INPH was assessed through clinical and radiological examinations in patients with a transient ischemic attack (TIA), consecutively admitted to the same hospital during 2006-2008. In the second cohort, VRFs, vascular disease and QoL were analysed in INPH patients consecutively shunted 2008-2010 in five out of six neurosurgical centres in Sweden. Patients remaining after inclusion (n=176, within the age-span 60-85 years and not having dementia) were compared to population-based age- and gender-matched controls (n=368, same inclusion criteria as for the INPH patients). Assessed VRFs were: hypertension, diabetes, obesity, hyperlipidemia, psychosocial factors (stress and depression), smoking, alcohol intake, physical activity and, dietary pattern. Cardiovascular, cerebrovascular and peripheral vascular disease as well as QoL were also assessed. Parameters were assessed through questionnaires, clinical examinations, measurements, ECG and, blood samples. In the first cohort, 4% of the TIA patients had clinically and radiologically verified INPH. In the second cohort, VRFs were overrepresented among the INPH patients compared with the controls. The VRFs independently associated with INPH were: hyperlipidemia (Odds ratio (OR): 2.4, 95%CI: 1.4-4.0), diabetes (OR: 2.2, 95%CI: 1.2-3.9), obesity (OR: 5.4, 95%CI: 2.5-11.8) and, psychosocial factors (OR: 5.3, 95%CI: 3.2-8.9). When adding the VRFs that were overrepresented in INPH, although not independently (physical inactivity and hypertension), these six VRFs accounted for 24% of the INPH cases in the elderly population (population attributable risk %: 24). Depression was overrepresented in shunted INPH patients compared to the controls (46% vs. 13%, p<0.001) and the main predictor for low QoL was a coexisting depression (p<0.001). In conclusion, the results of the INPH-CRasH study are consistent with a vascular pathophysiological component of INPH and indicate that INPH may be subgroup of vascular dementia. In clinical care and research, a complete risk factor analysis as well as screening for depression and a measurement for quality of life should be included in the work-up of INPH patients. The effect of targeted interventions against modifiable VRFs and anti-depressant treatment in INPH patients should be evaluated. / Idiopatisk normaltryckshydrocefalus (INPH, från engelskans ”idiopathic normal pressure hydrocephalus”) är en neurokirurgiskt behandlingsbar demens. Behandlingen är att operera in en shunt som dränerar cerebrospinalvätska från ventriklarna. Det har föreslagits att INPH skulle kunna orsakas av liknande patofysiologiska mekanismer som vid cerebrovaskulär sjukdom, men den vaskulära riskfaktorprofilen hos INPH-patienter har aldrig undersökts i en modern epidemiologisk studie. De kognitiva symtomen vid INPH påminner om symtomen vid depression, men prevalensen av depression hos INPH-patienter är okänd. Få studier undersöker hur shuntning påverkar livskvalitet och ingen studie har undersökt hur komorbiditet påverkar livskvaliteten vid INPH. Syftet med den här avhandlingen var att undersöka den vaskulära riskfaktorprofilen hos INPH-patienter samt att utforska hypotesen att INPH skulle kunna vara en undergrupp till vaskulär demens. Ytterligare ett syfte med avhandlingen var att undersöka hur många INPH-patienter som har depression samt undersöka hur shunting och komorbiditet påverkar livskvalitet vid INPH. I den första kohorten undersöktes kliniska och radiologiska fynd som tydde på INPH hos de patienter som blivit diagnostiserade med en TIA (från engelskans: transient ischemic attack) 2006-2008 på Norrlands Universitetssjukhus i Umeå. I den andra kohorten undersöktes konsekutivt shuntade INPH-patienter 2008-2010 från fem av sex neurokirurgiska kliniker i Sverige. De patienter som inkluderades i studien (n=176, ålder: 60-85 år, ej dementa) jämfördes med köns- och åldersmatchade kontroller från normalpopulationen (n=368, samma inklusionskriterier som för INPH-patienterna). De riskfaktorer som undersöktes var: hypertension, hyperlipidemi, diabetes, fetma, psykosociala faktorer (stress och depression), rökning, alkohol, fysisk aktivitet och diet. Även kardiovaskulära och cerebrovaskulära sjukdomar undersöktes, liksom perifer vaskulär sjukdom samt livskvalitet. Datainsamling skedde genom frågeformulär, kliniska undersökningar, mätningar, EKG och blodprov. I den första kohorten hade 4% av TIA-patienterna kliniskt och radiologiskt verifierad INPH. I den andra kohorten var vaskulära riskfaktorer överrepresenterade hos INPH-patienterna jämfört med iv normalpopulationen. Hyperlipidemi (OR: 2.4, 95%CI: 1.4-4.0), diabetes (OR: 2.2, 95%CI: 1.2-3.9), fetma (OR: 5.4, 95%CI: 2.5-11.8) och psykosociala faktorer (OR: 5.3, 95%CI: 3.2-8.9) var associerade med INPH oberoende av kön, ålder och de andra riskfaktorerna. Hypertension och fysisk inaktivitet var också associerade med INPH, dock inte oberoende av övriga riskfaktorer. Sammanlagd PAR% (från engelskans: population attributable risk %) för de här sex riskfaktorerna var 24%. INPH-patienterna hade depression i högre utsträckning än kontrollerna (46% vs. 13%, p<0.001), och depression var den viktigaste prediktorn för låg livskvalitet. Resultaten tyder på att vaskulär sjukdom och vaskulära riskfaktorer är involverade i den patofysiologiska mekanismen vid INPH. INPH kan vara en undergrupp till vaskulär demens. En fullständig riskfaktoranalys och screening för depression bör ingå i den preoperativa utvärderingen såväl som i forskning på INPH-patienter, och ett mått på livskvalitet bör införas. Effekten av riktade insatser mot såväl vaskulära riskfaktorer som depression vid INPH bör utvärderas.

Objectifs: Évaluer l’association entre l’exposition récréative, professionnelle et globale au soleil et le risque de cancer de la prostate (CaP). Méthodes: Dans le contexte d’une étude cas-témoins sur le CaP menée à Montréal, Canada, des entrevues ont été complétées auprès de 1371 cas incidents de CaP diagnostiqués en 2005-2009, et 1479 témoins de la population générale. Des questionnaires détaillés ont permis d’obtenir de l’information sur la fréquence et la durée de participation à toute activité extérieure lors des loisirs durant l’âge adulte, ainsi qu’une description de chaque emploi tenu au cours de la vie. Une matrice emploi-exposition canadienne a été appliquée à chaque emploi afin d’assigner un niveau d’exposition professionnelle au soleil. Des indices cumulatifs de l’exposition au soleil basés sur le nombre d’événements récréatifs, la durée d’exposition professionnelle, ainsi qu’un indice d’exposition global ont été développés. La régression logistique a été utilisée pour estimer l’association entre chaque indice d’exposition et le CaP, en ajustant pour des variables de confusion potentielles. Résultats: Globalement, il n’y avait pas d’association entre chacun des indices d’exposition et le risque de CaP. Certaines tendances en accord avec un risque légèrement plus faible chez les hommes exposés au soleil ont été observées mais les résultats n’étaient pas statistiquement significatifs et il n’y avait pas de relation dose-réponse. Conclusion: Notre étude apporte peu de soutien à l’hypothèse d’une association entre l’exposition au soleil et le risque de développer un cancer de la prostate. / Objectives: To investigate the association between sunlight exposure during leisure time, at work and globally, and prostate cancer (PCa) risk. Methods: In the context of a case-control study conducted in Montreal, Canada, interviews were conducted with 1371 incident PCa cases diagnosed between 2005 and 2009, and 1479 population controls. Detailed questionnaires were used to elicit the frequency and duration of engagement in any outdoor recreational activity during adulthood, as well as a description of each job held over the lifetime. A Canadian job-exposure matrix was applied to attribute a sunlight exposure level to each job. Cumulative indices of sunlight exposure were developed based on the number of outdoor leisure-time events, the duration of occupational exposure, separately and combined. Logistic regression was used to estimate the association between each sunlight exposure index and PCa, adjusting for potential confounding factors. Results: As a whole, there was no association between any of the exposure indices, and PCa risk, or PCa aggressiveness. Some trends for slightly lower PCa risks among men exposed to sunlight were observed, but results were not statistically significant and there was no dose-response pattern. Conclusion: Our findings provide little evidence for an association between sunlight exposure during adulthood and prostate cancer development.

Recent technological advances and remarkable successes have led to genome-wide association studies (GWAS) becoming a tool of choice for investigating the genetic basis of common complex human diseases. These studies typically involve samples from thousands of individuals, scanning their DNA at up to a million loci along the genome to discover genetic variants that affect disease risk. Hundreds of such variants are now known for common diseases, nearly all discovered by GWAS over the last three years. As a result, many new studies are planned for the future or are already underway. In this thesis, I present analysis results from actual studies and some developments in theory and methodology. The Wellcome Trust Case Control Consortium (WTCCC) published one of the first large-scale GWAS in 2007. I describe my contribution to this study and present the results from some of my follow-up analyses. I also present results from a GWAS of a bipolar disorder sub-phenotype, and a recent and on-going fine mapping experiment. Building on methods developed as part of the WTCCC, I describe a Bayesian approach to GWAS analysis and compare it to widely used frequentist approaches. I do so both theoretically, by interpreting each approach from the perspective of the other, and empirically, by comparing their performance in the context of replicated GWAS findings. I discuss the implications of these comparisons on the interpretation and analysis of GWAS generally, highlighting the advantages of the Bayesian approach. Finally, I examine the effect of linkage disequilibrium on the detection and estimation of various types of genetic effects, particularly non-additive effects. I derive a theoretical result showing how the power to detect a departure from an additive model at a marker locus decays faster than the power to detect an association.

Objectif : Examiner l’association entre la consommation alimentaire de caroténoïdes (β-carotène, α-carotène, β-cryptoxanthine, lutéine/zéaxanthine, lycopène) et de vitamine C et le risque de cancer du poumon, selon le sexe, l’intensité de tabagisme et le sous-type histologique de la tumeur. Méthodes : Les données proviennent d’une étude cas-témoins menée à Montréal, Canada. Des entrevues ont été effectuées auprès de 1 105 cas incidents de cancer du poumon et 1 449 témoins issus de la population générale. Leur fréquence de consommation moyenne de 49 fruits et légumes deux ans auparavant a été convertie en apports en antioxydants. Les rapports de cotes (RC) et les intervalles de confiance (IC) à 95% caractérisant l’association entre les antioxydants et le risque de cancer du poumon ont été estimés à l’aide de modèles de régression logistique et polytomée, en tenant compte de facteurs de confusion potentiels. Résultats : Une consommation élevée en antioxydants était généralement associée à une diminution du risque de cancer du poumon de l’ordre de 30%. Un effet protecteur a été observé chez les hommes et les femmes, pour les non fumeurs, les fumeurs quelque soit l’intensité de tabagisme, ainsi que pour les carcinomes à petites cellules, épidermoïde et l’adénocarcinome. Conclusions : Plusieurs antioxydants alimentaires protégeraient du cancer du poumon. Les efforts de prévention bénéficieraient de cibler la promotion de la consommation de fruits et de légumes riches en caroténoïdes et en vitamine C. / Objective: To investigate the association between dietary intake of carotenoids (β-carotene, α-carotene, β-cryptoxanthin, lutein/zeaxanthin and lycopene) and vitamin C, and risk of lung cancer according to sex, smoking intensity and tumor histological subtype. Methods: In the course of a case-control study conducted in Montreal, Canada, in-person interviews elicited dietary data from 1,105 incident lung cancer cases and 1,449 population controls. Usual frequency of intake of 49 fruit and vegetables two years prior to diagnosis or interview was estimated and converted to antioxidant intakes. Odds ratios (OR) and 95% confidence intervals (CI) between intake variables and lung cancer were estimated using logistic and polytomous regression models, adjusting for potential confounding factors. Results: High intakes of antioxidants were generally associated with some 30% reduction in lung cancer risk. A protective effect was observed among men and women, among never smokers, smokers regardless of intensity, and for small cell carcinoma, squamous cell carcinoma and adenocarcinoma. Conclusions: Results from this study suggest several dietary antioxidants may protect against lung cancer. Prevention programs should promote increased intakes of fruit and vegetables rich in carotenoids and vitamin C.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.