Genetic studies of lung disease in Cystic Fibrosis are faced with the challenge of identifying a severity measure that accounts for chronic disease progression and mortality attrition. Further, combining analyses across studies requires common phenotypes that are robust to study design and patient ascertainment. This thesis uses data from the North American Cystic Fibrosis Modifier Consortium (Canadian Consortium for CF Genetic Studies (CGS), Johns Hopkins University Twins and Siblings Study (TSS), and University of North Carolina/Case Western Reserve University Gene Modifier Study (GMS)), to calculate two novel phenotypes using age-specific CF percentile values of FEV1 (Forced Expiratory Volume in 1 second), with adjustment for CF age-specific mortality. The normalized residual, mortality adjusted (NoRMA) was designed for population based samples, while KNoRMA, using Kulich percentiles, is robust to sample ascertainment; both account for the effects of age-related disease progression and mortality attrition. NoRMA was computed for 2122 patients representing the Canadian CF population. KNoRMA was computed for these 2122 patients and also 1137 extreme phenotype patients in the GMS study and 1323 patients from multiple CF sib families in the TSS study. Phenotype was distributed in all three samples in a manner consistent with ascertainment differences, reflecting the lung disease severity of each individual in the underlying population. The new phenotype was highly correlated with the previously recommended mixed model phenotype1; 2, but computationally much easier and suited to studies with limited follow up time. As an example of its use, KNoRMA was used to test the association between locus variants in a previously published candidate gene, Transforming Growth Factor β1(TGFβ1), and lung function in CF, in an attempt to provide insight into discrepant results in the literature. A disease progression and mortality adjusted phenotype reduces the need for stratification or additional covariates, increasing statistical power and avoiding possible interpolation distortions.

Background: The relationship between average alcohol consumption and heart disease is well researched, showing a substantial cardioprotective association. This dissertation examined the epidemiological evidence for an effect of heavy episodic drinking (HED) over and above the effect of average alcohol consumption on heart disease. Methods: Electronic databases were systematically searched for epidemiological studies on the effect of HED on heart disease and identified articles were quantitatively summarized in a meta-analysis. Meta-regression models were used to examine the effect of characteristics of primary studies. Using individual-level data, semi-parametric Cox regression models were used to investigate HED exposure within narrow categories of average alcohol consumption in a US national population sample (n = 9,937) in relation to heart disease mortality in an 11-22 year follow-up. Frequency of heavy drinking episodes was used to identify latent classes of drinking history using growth mixture modeling in a sub-sample of this US cohort. Retrieved classes were used as independent variables in Cox regression models with heart disease mortality as the outcome event. Results: A pooled relative risk of 1.45 (95% confidence interval (CI): 1.24-1.70) for HED compared with non-HED drinkers with average alcohol consumption between 0.1-60 g/day was derived in a meta-analysis. A strong and consistent association with HED was found among current drinkers consuming an average of 1-2 drinks per day in the US cohort. There was no evidence of increased heart disease mortality resulting from the frequency of heavy drinking episodes before the age of forty. Conclusions: There is reasonable and consistent evidence for an association of HED and heart disease in current drinkers, negating any beneficial effect from alcohol consumption on heart health. History of frequency of heavy drinking episodes, however, showed no evidence for such an effect modification.

Type 2 Diabetes Mellitus (T2DM) prevalence is increasing globally with roughly 2.4 million people currently living with this condition in Canada. T2DM occurs more commonly in non-European ethnoracial groups, however the distribution of risk by age, sex, ethnicity and immigration status in Canada are not completely understood. The purpose of this thesis is to investigate the epidemiology of diabetes in an immigrant, multi-ethnic population using linked immigration and health data for the province of Ontario. The ultimate goal of this work is to generate information that can be used to design appropriate and effective targeted programs for diabetes prevention, management and control in order to reduce inequities in health. The principal findings of this work indicate that: 1) South Asians had a three-fold higher risk for developing diabetes as compared with people of European ethnicity and this disparity in risk was evident at a very young age; 2) The young age at diabetes onset experienced by many of our high-risk ethnic groups, including South Asians and people of African and Middle Eastern descent, suggest that in order to capture an equivalent risk of disease, screening may be recommended up to 15 years earlier in these groups – which is not reflected in current screening guidelines; 3) Contrary to patterns seen in Western European populations, women belonging to many high–risk ethnicities had equivalent or, in some cases, higher risk than men; 4) Risk varied substantially across country and region of birth making broad definitions of race or ethnicity (eg. ‘Asian’ or ‘Black’) inappropriate. These findings emphasize the heterogeneity of risk experienced by different ethnoracial populations in Canada and suggest that targeted primary prevention programs aimed at young adults and adolescents belonging to high-risk ethnic groups may be warranted. In addition, screening guidelines may need to be updated to reflect the younger age at onset in these populations. Further research is necessary to identify culturally appropriate and effective programs to reduce diabetes risk and associated health problems in these populations.

Background: The causes of trends in asthma incidence are not fully understood. Objectives: This study examined trends in age and severity at asthma diagnosis for Ontario children. Methods: Multiple birth cohorts of Ontario children between 1992-2000 were created using health administrative data. Descriptive statistics and multivariable logistic regression examined changes in age and severity of asthma at diagnosis over time. Results: Age at asthma diagnosis decreased (p<0.0001) with a higher relative risk of asthma in children under age three (RR=1.5, 95% CI:1.47, 1.54). Predictors of asthma diagnosis before three included male sex, lower income quintile, and maternal asthma. ‘Severe onset asthma’ increased over time (p<0.0001), its predictors being male sex, lower income quintile, rural residence, comorbidity, low birth weight and age less than three. Conclusions: Observed trends in asthma incidence are not confined to mild disease and are secondary to variations in asthma rates in children under age three.

The main objective of the thesis was to compare two methods of calculating adjusted attributable fractions and deaths as applied to smoking exposure and four health outcomes, lung cancer, ischaemic heart disease, chronic obstructive pulmonary disease, and cerebrovascular disease, for Canadians 30 years or older in the year 2005. An additional objective was to calculate variance estimates for the evaluation of precision. Such estimates have not been published for Canada to date. Attributable fractions were calculated using the fully adjusted method and the partial adjustment method. This method requires confounder strata specific (stratified) estimates of relative risk, along with accompanying estimates of variance. These estimates have not previously been published, and were derived from the Cancer Prevention Study II cohort. Estimates of the prevalence of smoking in Canada were obtained from the Canadian Community Health Survey 2005. Variance estimates were calculated using a Monte Carlo simulation. The fully adjusted method produced smaller attributable fractions in each of the eight disease-sex-specific categories than the partially adjusted method. This suggests an upwards bias when using the partial adjustment method in the attributable fraction for the relationship between cigarette smoking and cause-specific mortality in Canadian men and women. Summed across both sexes and the four smoking related causes of death, the number of deaths attributable to smoking was estimated to be 25,684 using the fully adjusted method and 28,466 using the partial adjustment method, an upward bias of over ten percent, or 2,782 deaths. It is desirable, theoretically, to use methods which can fully adjust for the effect of confounding and effect modification. Given the large datasets required and access to original data, using these methods may not be feasible for some who would wish to do so.

Occupational skin and respiratory symptoms, and disease, are common problems. Workers can develop new disease or aggravate existing disease as a result of exposures at work. Many workers are exposed to chemicals that can cause both respiratory and skin responses and there is evidence that some workers experience symptoms in both systems. There is also evidence that skin exposure may lead to sensitization and the development of respiratory disease. There is very little research that has examined both airborne and skin exposures together with lung and skin outcomes. The purpose of this thesis was to further investigate the relationships between occupational exposures, skin symptoms and disease, and respiratory symptoms and disease. Four studies were undertaken to improve our understanding of these complex relationships. Results from a study of clinical patch test data determined that seven of the ten most common occupational contact allergens are also capable of causing occupational asthma and that these common occupational exposures may not be recognized as sensitizers in common reference materials. Exposure-response relationships for skin symptoms were modeled in bakery workers and auto body shop workers using historical data; significant exposure-response relationships were found for auto body workers. In two separate studies of concurrent skin and respiratory symptoms, workers did report concurrent skin and respiratory symptoms. In predictive models, subjects reporting a history of eczema were more likely to report concurrent skin and respiratory symptoms. Overall, the results from this thesis provide more evidence that the skin and respiratory systems are associated. This body of work suggests that: (1) several common occupational exposures can cause disease in both the skin and respiratory system; (2) a portion of workers report both skin and respiratory symptoms; and (3) exposure-response relationships do exist for skin symptoms, both work-related and non-work-related. Future studies need to gather detailed information about exposure and response in both systems in order to better determine the role of exposure(s) in the development of skin and respiratory symptoms. Improved understanding of these relationships will allow for more targeted and effective exposure prevention strategies and will ultimately reduce the burden of occupational disease.

Background: Serum estradiol levels decline after menopause and the effect on mental health-related quality of life (MHR-QOL) is unclear. Objective: To determine if there is an association between endogenous serum estradiol levels and MHR-QOL in healthy postmenopausal women. Methods: This cross-sectional study used baseline Canadian data from the Mammary Prevention.3 trial. Serum estradiol was measured with liquid chromatography-tandem mass spectrometry. Outcomes for MHR-QOL were the Medical Outcomes 36-Item Short Form Health Survey (SF-36) Mental Health Inventory-5 (MHI-5), Mental Component Summary (MCS), and the Menopause-Specific Quality of Life Questionnaire (MENQOL)-psychosocial domain. Results: There were no statistically significant associations between estradiol levels and MHR-QOL in univariate analyses (n=455). Multivariable linear regression predicted statistically significant differences in MCS (R2=0.10, P=0.03) and MENQOL-psychosocial domain (R2=0.10, P=0.04), however estradiol was not a significant predictor. Conclusions: This study did not find a statistically significant association between endogenous serum estradiol levels and MHR-QOL in healthy postmenopausal women.

Contexte: Depuis plusieurs années, la fréquence et l’amplitude des températures chaudes extrêmes ont augmenté partout dans le monde. Ces évènements de chaleur extrême ont d’importants effets sur la santé. Pour ne citer qu’un exemple, la canicule de l’été 2003 en Europe a causé plus de 70 000 décès supplémentaires. La pression sélective dans l’évolution humaine conduisant à l’adaptation au climat a été mise en évidence par l’étude de Wyckelsma et al. rapportant que la variante de fonction rs1815739 au niveau du gène ACTN3 offre une résilience au froid grâce à la génération de chaleur musculaire. Cela confirme la pertinence de notre étude pangénomique visant à identifier des facteurs génétiques modifiant le risque de décès lors de températures chaudes extrêmes. Méthodes: Nous avons mené une étude cas-témoins à partir des données de la UK Biobank, qui compte environ 500 000 participants, pour recueillir les génotypes et les informations sur les lieux et dates de décès essentiels à notre étude. Les données environnementales de 2006 à 2020 ont été extraites du Met Office Integrated Data Archive System UK Hourly Temperature. Pour chaque décès, 4 individus témoins ont été sélectionnés aléatoirement parmi les individus de la UK Biobank ayant survécu jusqu’à la fin de la période de suivi. Nous avons effectué une analyse de régression logistique conditionnelle pour évaluer l'effet d'interaction entre les variants génétiques et l’exposition à la chaleur (dépassement du seuil de température extrême). Résultats: Nous avons recensé 27 284 décès entre 2006 et 2020, dont 939 sont survenus lors de dépassement du seuil de température extrême. Notre analyse pangénomique a identifié les variants génétiques rs1851364 sur le chromosome 3 et rs147921336 sur le chromosome 18 modifiant l’effet du dépassement du seuil de chaleur extrême sur l’augmentation du risque de décès (P<5x10-8). La présence d’au moins un allèle C pour chacun de ces variants génétiques double le risque de moralité lors d’une exposition à la chaleur. Conclusion: La génétique semble moduler l’effet des températures chaudes extrêmes associé à une augmentation du risque de décès chez l’humain. Ces résultats contribuent à une meilleure connaissance des mécanismes biologiques impliqués dans la susceptibilité aux températures extrêmes. / Context In recent years, the frequency and amplitude of extreme heat events have increased worldwide. These extreme heat events have significant effects on health. To take just one example, the 2003 heatwave in Europe caused over 70,000 additional deaths. The selective pressure in human evolution leading to climate adaptation was highlighted by the study of Wyckelsma et al. reporting that the function variant rs1815739 at the level of the ACTN3 gene offers resilience to cold through muscle heat generation. This supports the relevance of our genome-wide study aimed at identifying genetic factors modifying the risk of death during extreme hot temperatures. Methods We conducted a case-control study using data from the UK Biobank, which includes over 500,000 participants, to obtain information on genotypes and locations and dates of deaths required for our study. Environmental data from 2006 to 2020 were extracted from the Met Office Integrated Data Archive System UK Hourly Temperature. For each death, 4 control individuals were selected randomly from the pool of individuals in the UK Biobank who survived to the end of our study’s follow-up period. We performed a conditional logistic regression analysis to assess the interaction effect between genetic variants and heat exposure (exceeding the extreme temperature threshold). Results There were 27,284 deaths that occurred between 2006 and 2020, 939 of which occurred when the extreme temperature threshold was exceeded. Our genome-wide analysis identified that genetic variants rs1851364 on chromosome 3 and rs147921336 on chromosome 18 modified the effect of exceeding the extreme heat threshold on the increased risk of death (P<5x10-8 ). The presence of at least one allele C for each of these genetic variants doubles the risk of morality during heat exposure. Conclusion Genetics seems to modulate the effect of extreme hot temperatures associated with an increased risk of death in humans. These results contribute to a better understanding of the biological mechanisms involved in susceptibility to extreme temperatures.

Background and objectives Evidence suggests that circulating IGF-I levels increase mammographic density (a breast cancer risk factor) and breast cancer risk in premenopausal women. The objective of this thesis was to examine the association of genetic variation at the IGF1 gene with IGF-I concentration, mammographic density, breast cancer risk, and related anthropometric measures in premenopausal women. Methods Three IGF1 CA repeat polymorphisms (at the 5′ and 3′ ends, and in intron 2) were genotyped. A cross-sectional design was used to investigate their associations with IGF-I levels, mammographic density, BMI, weight, and height. Families from registries in Ontario and Australia were used to investigate associations with breast cancer risk and also BMI, weight and height. Results In the cross-sectional study, greater number of copies of the 5′ 19 allele were associated with lower circulating IGF-I levels. Greater number of 3′ 185 alleles were associated with greater percentage breast density, smaller amount of non-dense tissue, and lower BMI. Including BMI in regression models removed the association of the 3′ 185 allele with percentage breast density. In the family based study, nominally significant associations (5′ 21 allele, intron 2 212 allele, intron 2 216 allele) with breast cancer risk were observed, but significance was lost after multiple comparison adjustment. There was a stronger association between the intron 2 216 allele and risk under a recessive model, and 5′ allele groupings of length 18 to 20 and 20 or more repeats produced significant positive and negative associations respectively. These associations were not strongly supported in analyses stratified by registry. Results from the family based study did not support an association between genetic variation at IGF1 with BMI, weight or height. Conclusions No specific IGF1 variant influenced each of circulating IGF-I levels, mammographic density, and breast cancer risk. The failure to replicate the association of the 3′ 185 allele with BMI in the family based study suggests that the association of the 3′ 185 allele with percentage breast density is spurious, since this association was mediated through the relationship with BMI (suggesting IGF-I action on body fat). Evidence for an association between IGF1 and breast cancer risk was limited.

This dissertation involved linking a clinical cystic fibrosis (CF) data registry with administrative databases to evaluate clinical, demographic, and geographical predictors of hospitalization in CF patients living in Ontario over a 10 year period. In addition, this work assessed the ability of administrative data to identify individuals with CF using the clinical registry as the reference standard. Sex was an independent predictor of hospitalization rates for individuals with CF. Females had a significantly higher hospitalization rate compared to males even after adjusting for important clinical factors suggesting that this finding is not simply due to worse CF disease. In those between 7 and 19 years of age, the adjusted hospitalization rate was 38% higher in females (rate ratio[RR] 1.38, 95% confidence interval [CI] 1.11-1.73). Similarly in those over the age of 19, females had a 30% higher hospitalization rate compared to males (RR 1.30, 95% CI 1.06-1.59). Other significant predictors associated with higher hospitalization rates in both age groups were lower lung function, worse nutritional status, pancreatic insufficiency, and the presence of CF-related diabetes. The presence of Burkholderia cepacia complex in the sputum was a significant predictor in those over the age of 19 years (RR 1.54, 95% CI 1.26-1.89). Distance to CF centre, community size and socioeconomic status were not significant predictors of hospitalization rates in either age group. There was no significant trend in hospitalization rates over time once rates were adjusted for markers of disease severity (p=0.08). Comparing administrative data with the CF registry data, administrative data captured hospitalizations more comprehensively. Despite CF being a specific diagnosis, health administrative databases alone were insufficient to reliably and accurately identify individuals with CF unless they had been hospitalized. The reason for the gender disparity seen within this dissertation is likely multifactorial. There may be differences in outpatient management between the sexes, hormonal influences may modulate disease severity causing higher hospitalization rates, and patient and provider-level influences may affect the decision to hospitalize a patient. Further research is needed in this area to elucidate the factors contributing to this gender gap.