Aperçu

Résumé (français)

Veuillez noter que les résumés n'apparaissent que dans la langue de la publication et peuvent ne pas avoir de traduction.

Résumé (anglais)

National nutrition surveys are the cornerstones of nutritional surveillance for developing dietary guidelines and policies. Some recent studies have questioned the usefulness of nutrition surveys due to their methodological limitations. The overall goal of this thesis was to use the Canadian Community Health Survey (CCHS) 2.2 to address these limitations as the first step in population-based dietary pattern analysis, an essential component for development of evidence-based nutritional guidelines and policies. In the first study of this thesis, different methods for handling dietary misreporting were compared and “adjusting for misreporting bias” was identified as the most appropriate technique, which was used in all subsequent studies of this thesis. In the second study, we observed that closer adherence to the only Canadian a priori index, Health Canada’s Surveillance Tool Tier System (HCST) 2014, developed based on the Eating Well with Canada’s Food Guide 2007 (EWCFG) was associated with higher probability of meeting dietary reference intakes (DRI) for nutrients, even though it was not related to obesity risk. These findings were explained in the third study, where the strict focus of the EWCFG on single nutrients, rather than dietary patterns was identified as its main limitation. The first Canadian dietary pattern analyses using energy-based a priori (Study 4) and hybrid (Study 5) techniques were then conducted to address this limitation. Lack of adherence to the recommendations of 2015 Dietary Guidelines for Americans Adherence Index (DGAI), an a priori dietary quality index, and consumption of an energy-dense, high-fat and low-fiber dietary pattern derived from the weighted partial least squares, were associated with 2-3 times higher risk of obesity. Overall, studies in this thesis demonstrate that application of rigorous methodological techniques to survey data can enhance the usefulness of nutrition surveys for capturing the diet-disease relationships and for informing evidence-based national nutrition guidelines and policies.