"Next Generation" Approaches in Diet Pattern Analysis: Assessing the Impact of Different Statistical Methods and Physiological Intermediate Variables
Year: 2016
Type: dissertation
Abstract: The objective of this thesis was to compare dietary patterns (DPs) derived using principal component analysis (PCA), reduced rank regression (RRR) and partial least square (PLS) analysis as well as to evaluate how the selection of disease-related intermediary variables (DRIVs) alters the DPs produced. The association of DPs with type 2 diabetes mellitus (T2DM) was also explored across different statistical techniques and DRIVs. All analyses were conducted using the Insulin Resistance Atherosclerosis Study (IRAS), a cohort consisting of middle-aged adults initially free of T2DM. PCA, PLS and RRR and each a priori defined DRIVs (inflammatory biomarkers) each yielded distinctive sets of DPS; however, there were no significant differences in how the different methods or DRIVs predicted T2DM. This could be due to the limited number of T2DM converters, the duration of follow-up, the characteristics of the FFQ and/or the population under investigation.
Source: TSpace (University of Toronto)
Author: Robin Alyssa Glicksman
Language: English
Topic: Nutritional Studies and Diet
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W2602453591