Background: Canada relies heavily on family and friend caregivers to care for its increasing older population. However, many caregivers experience distress, which can jeopardize their ability to continue care provision and the health of their care-recipients. Objectives: This thesis research has three objectives: 1) Provide an overview of the burden and trajectories of caregiver distress in Ontario, and to compare differences in the experience of distress between caregivers of men and women; 2) Determine the association between caregiver distress and care-recipients' location of death; 3) Develop and test a comprehensive model of the relationships between caregiving factors, caregiver and care-recipient profiles, and caregiver distress. Methods: The main data source for objectives 1 and 2 is the Resident Assessment Instrument for Home Care, linked to multiple health administrative datasets at ICES. For objective 1, older (50+) community-dwelling adults in Ontario and their caregivers were examined. Descriptive analyses were performed to identify baseline and one-year change in caregivers' distress status, stratified by year of baseline assessment and care-recipients' level of care needs. Logistic regression was performed to identify the associations between care-recipients’ gender on caregiver distress. For objective 2, a retrospective cohort study of adult decedents in Ontario was conducted. Their caregivers' distress status within one-year of death was described, and logistic regression was performed to determine the association between caregiver distress and the odds of dying in non-palliative acute care. The data source for objective 3 is the General Social Survey-Caregiving & Care-receiving (cycle 26) conducted by Statistics Canada. The study population included respondents who have provided unpaid care within one year of survey. Exploratory factor analysis and structural equation modelling were performed to test a theoretical model of caregiver distress and its contributing factors and covariates. Results: From 2008 to 2015, there was a steady increase in prevalent, incident, and sustained caregiver distress. The increase was especially prominent among caregivers of individuals with lower care needs. Caregivers of older men were more likely to be distressed than caregivers of older women, largely because older men have greater health and functional impairments and required more care. Individuals cared for by distressed (vs. non-distressed) caregivers spent more time in non-palliative acute care in their last month of life and were more likely to die in non-palliative acute care. Receipt of home care and palliative home care greatly reduced the odds of dying in non-palliative acute care. Exploratory factor analysis established a well-fit model that represented caregiver distress and its five contributing factors: caregiving burden, disruptions of family and social life, caregiving history, caregiving network and support, and positive emotional experiences. Subsequent structural equation modelling found that disruptions of family and social life and positive emotional experiences had the largest associations with caregiver distress. Implications: Supporting older Canadians and their caregivers is a policy priority. This thesis research highlighted a rising trend of caregiver distress, identified risk and protective factors of distress, and determined the effect of caregiver distress on care-recipients' place of death. These findings can inform policy decisions and facilitate health systems planning. The risk factors found in this research can also be integrated into clinical assessments to identify caregivers at high risk of distress to provide them with timely and appropriate support. Enabling caregivers to provide quality care without being distressed is crucial for reducing cost incurred by the negative impacts of caregiving, ultimately reducing the overall healthcare cost of the aging population.