Grant: Master's Award: Canada Graduate Scholarships
McMaster University
$17,500.00
Exploring the Effect of Nutrition on Mortality and Cardiovascular Disease using Linked Population-based Dietary Data
FY 2018-19
Summary
Diet is a major determinant of health and is considered to be largely responsible for the reduction in cardiovascular disease (CVD) in industrialized countries since the 1960s. Dietary recommendations focus on elimination of trans-fat, reduction of saturated fatty acids (SAFAs), and increases in vegetable and fruit consumption. However, recently it has been highlighted that the evidence-base may not be very strong for these recommendations. For example, systematic reviews fail to demonstrate consistent relationships between SAFAs, CVD, and mortality. The most recently published large nutritional cohort study has identified carbohydrates as detrimental, while fats were protective against death and CVD. Then, a meta-analysis published this summer suggested a u-shaped relationship between carbohydrate and mortality, and that fats are only protective if they're plant-based. Given the inconsistencies, a reliance on predetermined nutrient categories may be problematic. I propose applying novel machine learning techniques to both identify the dietary components truly related to health, and to better characterize the complex relationship. I plan to use the Canadian Community Health Survey (CCHS), which contains the only detailed Canadian population-based dietary data available and has been linked to participants’ later healthcare utilization. This will allow development of a model to both predict CVD and mortality and determine their dietary risk factors, applying both traditional and ML techniques. This would be the first time that such a study has been performed in Canada, and could help identify whether ML may be useful. Ultimately, the project aims to better inform decision-makers, clinicians, and individuals about diet-related policy and behaviour impacts on health.
Department
Program
Master's Award: Canada Graduate Scholarships
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
406315
Name
Morgenstern Jason D
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Exploring the Effect of Nutrition on Mortality and Cardiovascular Disease using Linked Population-based Dietary Data
Programtype2
Studentship Programs
Orgnameinp2
McMaster University
Cihramount2
$17,500
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Biostatistics; Cardiovascular Disease; Diet; Epidemiology; Machine Learning; Nutritional Epidemiology; Population Health; Risk Prediction
Abstract
Diet is a major determinant of health and is considered to be largely responsible for the reduction in cardiovascular disease (CVD) in industrialized countries since the 1960s. Dietary recommendations focus on elimination of trans-fat, reduction of saturated fatty acids (SAFAs), and increases in vegetable and fruit consumption. However, recently it has been highlighted that the evidence-base may not be very strong for these recommendations. For example, systematic reviews fail to demonstrate consistent relationships between SAFAs, CVD, and mortality. The most recently published large nutritional cohort study has identified carbohydrates as detrimental, while fats were protective against death and CVD. Then, a meta-analysis published this summer suggested a u-shaped relationship between carbohydrate and mortality, and that fats are only protective if they're plant-based. Given the inconsistencies, a reliance on predetermined nutrient categories may be problematic. I propose applying novel machine learning techniques to both identify the dietary components truly related to health, and to better characterize the complex relationship. I plan to use the Canadian Community Health Survey (CCHS), which contains the only detailed Canadian population-based dietary data available and has been linked to participants’ later healthcare utilization. This will allow development of a model to both predict CVD and mortality and determine their dietary risk factors, applying both traditional and ML techniques. This would be the first time that such a study has been performed in Canada, and could help identify whether ML may be useful. Ultimately, the project aims to better inform decision-makers, clinicians, and individuals about diet-related policy and behaviour impacts on health.
Competitiondate
201812
Approvedterm2
1 yr 0 mth
Pinamesdelim
Morgenstern, Jason D
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $17,500~ $0~