Grant: Doctoral Research Award: Canada Graduate Scholarships
University of Toronto
$105,000.00
Developing Population-Based Risk Tools to Predict and Reduce Premature Mortality in Canadian Cities
EF 2019-20
Résumé
Premature mortality, defined as deaths before 75 years of age, is considered avoidable through appropriate prevention or timely treatment. In Canada, premature mortality rates vary across sex, socioeconomic status, and geography. In this study, I will develop a tool to predict premature mortality, which will support policy and health system decision-making. First, I will identify social, environmental, and clinical risk factors for premature mortality. I will then use these risk factors to create the prediction tool using machine learning. This research will identify which communities are at higher risk, highlighting prevention opportunities for diverse populations. As well, health planners will be able to use this tool to test the impact of health interventions aimed at reducing premature mortality rates. This tool will be capable of predicting the effects of health policies that affect over 21-million people living in urban areas, 60% of the Canadian population. This prediction tool will be the first of its kind and will be built specifically for a Canadian context. This prediction tool is novel because it will be capable of incorporating information on people's social and physical environments as well as their clinical history and health-related behaviours. Incorporating social and environmental data has two benefits: it helps the model make more accurate predictions and it integrates a broader focus on the determinants of health. Including social and environmental perspectives emphasizes prevention and equity, and it incorporates these perspectives into machine learning applications for population health. Creating this prediction tool requires developing new predictive methods that combine the fields of epidemiology and machine learning. By leveraging the increasing availability of health data and growing computational power, this research helps meet the growing demand for analytic, decision-support health tools.
Programme
Doctoral Research Award: Canada Graduate Scholarships
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
417003
Name
Pagalan Lief
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Population and Public Health
Theme2
Social / Cultural / Environmental / Population Health
Prcname2
Doctoral Research Awards - B
Projecttitle
Developing Population-Based Risk Tools to Predict and Reduce Premature Mortality in Canadian Cities
Programtype2
Studentship Programs
Orgnameinp2
University of Toronto
Cihramount2
$105,000
Cihrequipment2
$0
Programname2
Doctoral Research Award: Canada Graduate Scholarships
Keyworddelim
Environmental Health; Epidemiological Methods; Machine Learning; Modelling And Simulation; Premature Mortality; Risk Prediction
Abstract
Premature mortality, defined as deaths before 75 years of age, is considered avoidable through appropriate prevention or timely treatment. In Canada, premature mortality rates vary across sex, socioeconomic status, and geography. In this study, I will develop a tool to predict premature mortality, which will support policy and health system decision-making. First, I will identify social, environmental, and clinical risk factors for premature mortality. I will then use these risk factors to create the prediction tool using machine learning. This research will identify which communities are at higher risk, highlighting prevention opportunities for diverse populations. As well, health planners will be able to use this tool to test the impact of health interventions aimed at reducing premature mortality rates. This tool will be capable of predicting the effects of health policies that affect over 21-million people living in urban areas, 60% of the Canadian population. This prediction tool will be the first of its kind and will be built specifically for a Canadian context. This prediction tool is novel because it will be capable of incorporating information on people's social and physical environments as well as their clinical history and health-related behaviours. Incorporating social and environmental data has two benefits: it helps the model make more accurate predictions and it integrates a broader focus on the determinants of health. Including social and environmental perspectives emphasizes prevention and equity, and it incorporates these perspectives into machine learning applications for population health. Creating this prediction tool requires developing new predictive methods that combine the fields of epidemiology and machine learning. By leveraging the increasing availability of health data and growing computational power, this research helps meet the growing demand for analytic, decision-support health tools.
Competitiondate
201911
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Pagalan, Lief
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $105,000~ $0~