Thesis (Master, Community Health & Epidemiology) -- Queen's University, 2011. Includes bibliographical references (leaves 120-136)

More than half of all cases of breast cancer occur among women without any known risk factors. More research is needed on suspected risk factors in order to refine current breast cancer screening tools. The objectives of this thesis were: to determine the breast cancer risk associated with known risk factors (ethnicity, family history, breast biopsy, age at menarche, age at first birth, alcohol, HRT, and BMI), suspected risk factors (smoking, second-hand smoke exposure, smoked/grilled foods, and NSAID use), and to examine the above associations according to tumour receptor status, histologic grade, and menopausal status, with potential confounders also considered. This thesis project was conducted within the framework of the Molecular Epidemiology of Breast Cancer study, a case-control study of women in Vancouver, BC and Kingston, ON, with 1140 cases and 1169 controls recruited from 2005-2010. Information was collected from a detailed questionnaire. Cases and controls were similar in terms of age at menarche, age at first birth, smoking history, second-hand smoke exposure, lifetime smoked/grilled food consumption, HRT, and BMI. Among cases, there were significantly less Europeans and more Chinese, Japanese, and Filipino subjects compared to controls. Cases were more likely to have a first degree relative with breast cancer, as well as a previous benign breast biopsy. Alcohol consumption and past NSAID usage was higher among controls. The level of education completed was higher among controls. Cases were also more likely to be postmenopausal. Family history was associated with breast cancer risk (OR=1.59, CI=1.30-1.94), as was BMI (OR=1.28, CI=1.05-1.58 for overweight and OR=2.28, CI=1.35-3.86 for obese class II). Second-hand smoke was also found to be associated with breast cancer risk (OR=1.42, CI=1.02-1.97 for individuals with a less than 10 pack-year smoking history). Due to reduced sample size with stratification and marginally significant results, it is not possible to draw definitive conclusions regarding pathology sub-types. In summary, these results provide support for the association between several risk factors and breast cancer risk. More research is needed to ascertain how receptor status, histologic grade, and menopausal status affect these associations.