The ultimate success of bypass surgery depends on the construction of a technically perfect bypass graft to an appropriate coronary vessel using a conduit which will remain durable for the lifetime of the patient. This thesis explores methods to improve coronary surgery by enhancing intraoperative imaging and conduit selection in the operating room. It is known that technical errors in graft construction cause failure of up to 12% of coronary bypass grafts in the operating room. We performed investigations of a new technique of intraoperative fluorescence angiography using indocyanine green dye to determine graft patency. We developed optimal methods of obtaining images and preliminary investigations revealed the technique was highly reproducible. In a follow-up trial, we demonstrated that over 80% of technical errors which would otherwise have been missed were identifiable with indocyanine green angiography, while only 25% of these errors were identified by transit-time ultrasonic flow measurement, the current clinical standard. We also determine that coronary surgery with indocyanine green angiographic graft patency verification was associated with less perioperative myocardial injury than bypass surgery without graft patency assessment. The long term graft patency of saphenous vein grafts is sub-optimal, with over 40% of such grafts totally occluded and a further 30% significantly diseased at ten years. We attempted to improve these outcomes by increasing the use of arterial grafts, which are less prone to intimal hyperplasia. In a multicentre clinical trial, we demonstrated a 40% relative risk reduction in graft occlusion at one year when radial arteries were used as bypass conduits versus saphenous veins. We identified that women and patients with small coronary vessels maximally benefited from radial artery bypass grafts. Conversely, in settings of less severe target vessel stenosis or concomitant peripheral vascular disease, saphenous veins performed as well as radial arteries. We have demonstrated that high quality imaging to identify technical errors during the operation, increased use of radial artery grafts and careful consideration of individual patient and target vessel characteristics can all improve graft patency. Future studies will be aimed at identifying the role of intraoperative imaging and arterial grafting in improving long-term clinical outcomes.

Background: The use of meta-analysis to combine the results of observational studies is controversial. Despite its common use, methodological work in this area is lacking. Because of the diversity of study designs, exposure and outcome measures, and differential adjustment for confounding variables, the identification of sources of heterogeneity among study effect estimates is particularly important when combining data from observational studies. This thesis presents the results of a study that examines the relative ability of individual participant data (IPD) meta-analysis (which was considered a “gold standard”) and traditional aggregate data (AD) meta-analysis to identify sources of heterogeneity among studies examining mechanical exposure and low back pain (LBP) in workers. Materials and Methods: A systematic literature search was conducted to identify relevant articles. The corresponding author of each article was contacted to request their individual-level data. Because the outcome definitions and exposure measures were not uniform across studies, two sub-studies were conducted 1) to identify sets of outcome definitions that could be combined in a meta-analysis and 2) to develop methods to translate mechanical exposure onto a common metric. IPD analyses were conducted using generalized estimating equation (GEE) regression to identify variables that acted as strong confounders and effect modifiers. Traditional AD meta-analysis was also conducted and potential sources of heterogeneity were tested using meta-regression. Key Findings: (1) Overall, we found an association between both forces and postures on LBP, although the magnitude varied depending on the exposure-outcome combination. Among the outcomes, the ORs tended to be highest for sick leave due to LBP. (2) There was very little evidence of strong confounders in the relationship between mechanical exposure and LBP; thus differential adjustment for confounders in studies would not likely be an important source of heterogeneity in an AD meta-analysis. (3) AD meta-analysis was able to identify the same study-level effect modifiers as IPD meta-analysis, but did not consistently identify individual-level effect modifiers. Both individual-level characteristics (older age and being male), and study-level characteristics (population-based studies and self-reported mechanical exposure), were associated with an increased OR for many of the LBP outcome and mechanical exposure combinations. Conclusion: AD meta-analysis is likely sufficient to detect heterogeneity for study-level factors but is not sufficient to identify individual-level effect modifiers. When the primary source of evidence in a research area is observational studies and when there is controversy despite several systematic reviews, IPD meta-analysis can be used to better understand sources of heterogeneity and provide context

Rationale/Objective: Excess lung cancer risks associated with exposure to inhaled radon decay products among uranium miners has well been established. Although ingestion is also a potentially important route of exposure, the relationship between ingested radon decay products and gastrointestinal cancer risks are not well examined. The objective of this study is to determine the relationship between exposure to radon decay products and the incidence and mortality of gastrointestinal (esophagus, stomach, and colorectal) cancer among men employed as uranium miners in Ontario. Secondly, to determine whether the duration of exposure (dose rate), years since last exposure and age at first exposure modify these associations. Methods: A cohort of miners who had ever worked in an Ontario uranium mine between 1954 and 1996 was created using the Mining Master File and the National Dose Registry. Cumulative radon exposures measured in Working Level Months (WLM) were previously estimated for each miner. Cancer diagnoses (1964-2004) and cancer deaths (1954-2004) occurring in Ontario were determined by probabilistic record linkage with the Ontario Cancer Registry. To calculate person-years at risk, non-cancer deaths were also ascertained from the Ontario mortality file for the period between 1954 and 2004. Poisson regression methods for grouped data were used to estimate the relative risks (RR) and 95% Confidence Intervals (CI) by exposure level. Results/Conclusions: The final cohort consisted of 28,273 Ontario uranium miners. By the end of 2004, 34 miners had been diagnosed with esophageal cancer, 86 with stomach cancer, and 359 with colorectal cancer. There were 40 deaths due to esophageal cancer, 69 from stomach cancer, and 176 from colorectal cancer. When comparing the highest cumulative exposure category (>40 WLM) to the referent group (0 WLM), significant increases in both stomach (RRIncidence= 2.30, 95% CI;1.02-5.17 and RRMortality=2.90, 95% CI;1.11-7.63) and colorectal cancers (RRIncidence =1.56, 95% CI;1.07-2.27 and RRMortality =1.74, 95% CI;1.01-2.99) after adjusting for age at risk and period effects. However, no relationships were observed for esophageal cancer. Suggestive evidence of modifying effects of these associations by duration of employment (dose rate) and years since last exposure for colorectal cancer was also observed.

Striking differences exist between countries in the incidence of breast cancer, with rates higher in the West than in Asian countries. The causes of these differences are unknown, but because incidence rates change in migrants, they are thought to be due to lifestyle rather than genetic differences. The objective of this thesis was to compare established breast cancer risk factors, physical activity, and diet in three groups of postmenopausal women at substantially different risks of developing breast cancer – Caucasians (N = 413), Chinese born in the West or who migrated to the West before age 21 (N = 216), and recent Chinese migrants, 99% of whom coming from urban China (N = 421). In this cross-sectional study, information on risk factors and diet were collected by telephone, and physical activity and anthropometric data were obtained at a home visit. Compared to Caucasians, recent Chinese migrants weighed on average 14 kg less, were 6 cm shorter, had menarche a year later, were more often parous, and less often had a family history of breast cancer or a benign breast biopsy. Estimating 5-year absolute breast cancer risks using the Gail Model showed that risk estimates in Caucasians would be reduced by only 11% if they had the risk factor profile of recent Chinese migrants for the variables in the Gail Model. Compared to Caucasians, recent Chinese migrants had lower average total physical activity over lifetime, and also spent less time on moderate- and vigorous-intensity activity. Compared to Caucasians, recent Chinese migrants consumed per day on average 175 fewer calories, 6 more grams of energy-adjusted protein, 16 more grams of energy-adjusted carbohydrates, and 5 fewer grams of energy-adjusted fat. Also, recent Chinese migrants consumed higher amounts of grains, fruits, vegetables, fish, and soy products, and lower amounts of alcohol, meat, dairy products, and sweets than Caucasians. Western born Chinese and early Chinese migrants had values intermediate between the other two groups for most of the variables. These results suggest that in addition to the established risk factors, some dietary factors may also contribute to the lower breast cancer risk in urban Chinese women.

This thesis comprises a detailed analysis of sexual risk behavior among female sex workers participating in a prospective cohort study in Mombasa, Kenya between 1993 and 2007. To determine whether high-risk behavior has decreased over time, I examined trends in and associations with condom use and partner numbers at enrolment and over follow-up using multinomial logistic regression. While condom use increased among women enrolling into the cohort, women reduced partner numbers, rather than increasing condom use, over cohort visits. Workplace, charge for sex, duration of sex work, alcohol use, pregnancy and illness were all predictors of condom use. To evaluate the extent to which HIV risk estimates were affected by loss to follow-up, I investigated associations between sexual risk behavior, loss to follow-up, and HIV acquisition, using competing risks regression. Women reporting unprotected sex with multiple partners had the highest risk for HIV infection, and were also most likely to remain in the cohort. Finally, I used Andersen-Gill modeling to assess the impact of sexual risk behaviors on acquisition of sexually transmitted infections (STI) including gonorrhea, non-specific cervicitis, and trichomoniasis. While incident gonorrhea was closely associated with recent sexual risk behavior, incident trichomoniasis was not. Both conditions had high hazards for recurring in a 90-day window after a prior diagnosis. Non-specific cervicitis was demonstrated to be a chronic, relapsing condition associated with protected sex with multiple partners (possibly due to more frequent condom use) and with known biologic risk factors (i.e., pregnancy, hormonal contraceptive use, cervical ectopy, and genital ulcer disease). Overall, these analyses have led to a better understanding of how different sexual behavior patterns are associated with adverse outcomes, including HIV and STI acquisition, and identified specific factors associated with high-risk sexual behavior that may be amenable to intervention.

Violence and traumatic brain injury (TBI) are two major public health concerns. This thesis is comprised of three different research topics; the epidemiology of intentional TBI in Ontario, discharge against medical advice (DAMA) as an undesirable outcome of acute stage, and functional changes after receiving rehabilitation care. To study these areas, three different datasets from the Canadian Institute for Health Information (CIHI) were used. The first epidemiological study on intentional TBI identified 1,409 (8.0%) intentional TBIs and 16,211 (92.0%) unintentional TBIs. Of the intentional TBIs, 389 (27.6%) were self-inflicted TBI (Si-TBI) and 1,020 (72.4%) were other-inflicted TBI (Oi-TBI). The most common causes of Si-TBI were “jumping from high places” and “firearms”. Major causes of Oi-TBI were ‘fight and brawl” and “struck by objects”. Si-TBI was associated with younger age, female gender, and having a history of alcohol/drug abuse. Oi-TBI was also associated with younger age and having an alcohol/drug abuse history and also with male gender. The second study on discharge against medical advice found that 446 (2.84%) TBI patients left hospitals without medical advice. DAMA was significantly associated with intentional injuries in those with self-inflicted TBI and other-inflicted TBI. DAMA was also associated with younger age and a history of alcohol/drug abuse. Using univariate analyses, the third study found that people with intentional TBI had significantly lower FIM gains in the motor area and significantly lower relative function gains (as measured by Montebello Rehabilitation Factor Score) in the cognitive area. Multivariate analyses of the same data showed that intentional TBI was also associated with lower cognitive relative gains, while controlling for age, gender, alcohol/abuse history, and other demographic and clinical variables. Persons with intentional TBI were found to be less likely to be discharged home, controlling for other relevant confounders. In conclusion, a person who has been injured due to assault or suicidal attempt may need more individualized care as they may be at greater risk for adverse rehabilitation outcomes. These findings regarding people with intentional TBI provide a basis for enhancing efforts on prevention of violence-related TBI and DAMA, and also for improving rehabilitation programs and discharge plans for this vulnerable population.

Introduction: Trauma centers (TC) are the only institutions with resources to manage patients with severe traumatic brain injury (TBI). We chose to examine potential barriers to access to TC care for TBI patients. Methods: Administrative datasets were used to evaluate access to TC among patients with severe TBI. We examined triage practices of EMS in TBI. Finally, we analyzed surveys to capture the beliefs, perceptions and knowledge of ED physicians with respect to TBI. Results: 57% of patients in Ontario had any access to a TC following TBI. Of patients who had potential access to a TC from the scene of injury as defined by pre-hospital triage guidelines, 60% of patients were undertriaged. Challenges that ED physicians faced with managing TBI, included lack of beds at TC and difficulty attaining transport resources. Conclusion: Access to TC care for patients with TBI is impeded by patient and system level factors.

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.

The incidence of differentiated thyroid cancer (DTC) is rising, but controversy exists in many aspects of its treatment. This study described the change in incidence of DTC in Ontario, variations in management including extent of thyroidectomy and the influence of provider volume, and the impact of these parameters on recurrence and thyroid cancer-specific death (TCSD). A population-based study identified all new cases of DTC between 1992-2007. The incidence of DTC increased dramatically (annual percentage change 7.6%). Linkage to administrative databases revealed that extent of thyroidectomy is influenced by various factors including patient gender, age, year of diagnosis, surgeon specialty, and hospital setting, but not provider volume. Total thyroidectomy is associated with a lower recurrence rate. There is a significant association between provider volume and recurrence, with lower volume surgeons having a higher recurrence risk. Extent of thyroidectomy and provider volume did not influence TCSD. Such variations in management may lead to disparities in health outcomes.

The incidence of differentiated thyroid cancer (DTC) is rising, but controversy exists in many aspects of its treatment. This study described the change in incidence of DTC in Ontario, variations in management including extent of thyroidectomy and the influence of provider volume, and the impact of these parameters on recurrence and thyroid cancer-specific death (TCSD). A population-based study identified all new cases of DTC between 1992-2007. The incidence of DTC increased dramatically (annual percentage change 7.6%). Linkage to administrative databases revealed that extent of thyroidectomy is influenced by various factors including patient gender, age, year of diagnosis, surgeon specialty, and hospital setting, but not provider volume. Total thyroidectomy is associated with a lower recurrence rate. There is a significant association between provider volume and recurrence, with lower volume surgeons having a higher recurrence risk. Extent of thyroidectomy and provider volume did not influence TCSD. Such variations in management may lead to disparities in health outcomes.