Overview

Abstract (English)

This thesis investigates the effects of immigration status, country of origin, and duration in Canada on three main health outcomes, namely health care utilization, occurrences of hypertension and heart disease, and body mass index. The first two chapters are cross-sectional studies that utilize data derived from linked national health survey and Ontario databases, whereas the third chapter is a longitudinal study which draws data from the longitudinal National Population Health Survey (NPHS). The first chapter examines the role of immigration status and country of origin in explaining the use of three types of health services: primary care physicians, specialists, and hospitals. Negative binomial regressions are used to assess the associations of immigration status, country of origin, and the utilization of physician visits, whereas probit regressions are used to quantify the associations of immigration status, country of origin, and the likelihood of being hospitalized. The findings suggest that immigrants, especially those who are male and have low educational attainment, use more primary care physicians than comparable non-immigrants. Immigrants are found to use fewer expensive health services, i.e. specialist and hospital care, compared to Canadian-born residents. Likewise, immigrants from non-traditional source countries, i.e. those without a regular medical doctor, make even fewer visits to specialists than do those who came from traditional source countries. The second chapter investigates the associations of immigration status, occurrence of hypertension, and occurrence of heart disease. The logistic regressions and propensity score matching are performed to adjust for potential conf ounders. findings drawn from both the logistic regressions and propensity matching technique indicate that immigrants have comparable odds of hypertension and heart disease to those of Canadian-born residents after adjusting for other factors. Notably, the results of the exact propensity score matching reveal that the odds of hypertension are significantly higher among immigrant than non-immigrant men aged 42-51 years, whereas the odds of heart disease are significantly lower in immigrants than in Canadian-born residents aged 52-65 years. The third chapter uses the longitudinal files of the NPHS to examine the effects of time since arrival in Canada on the change in BMI over the 14-year period. The fixed effect models are used to assess the effects of time since arrival in Canada, a proxy of acculturation, on the change in BMI while controlling for the effect of arrival cohort. Moreover, a series of panel data analyses are used to address the problems commonly found in observational studies, namely sample selection bias, endogeneity, and reporting bias. This chapter shows that, holding other factors constant, an additional year in Canada leads to a 0.14% increase in an individual’s BMI. This association is found to be more pronounced for women than men and for married than non-married individuals. The effect of time since arrival in Canada on the change in BMI is reduced to 0.07% after adjusting for sample selection bias, suggesting that by ignoring the sample selection issue, the effects of time since arrival in Canada on the change in BMI may be overestimated.

Abstract (French)

Please note that abstracts only appear in the language of the publication and might not have a translation.