Cross-sectional associations between empirically-derived dietary patterns and indicators of disease risk among university students

Abstract

The transition from adolescence to adulthood is a unique period during which lifelong dietary habits are shaped. Dietary patterns (DPs) among young adults attending college have not been adequately described, and associations between DPs and indicators of disease risk are not well understood in this age group. Dietary data were collected from undergraduates participating in the Tufts Longitudinal Health Study (TLHS; 1998–2007) by Food Frequency Questionnaire (FFQ; n = 1323). DPs were derived using principal components analysis with varimax rotation. Scree plots; eigenvalues; factor loadings; and previous studies were used to determine and label the DPs retained. Cross-sectional relationships between DP scores and anthropometric measures (percent body fat (PBF) and (BMI) and lipid biomarkers (total; HDL and LDL cholesterol; and triglycerides) were assessed with multivariable regression models; adjusted for demographics; physical activity; smoking; intention to gain/lose weight; and total energy intake. Effect modification by sex was tested. Three DPs were identified: Prudent;Western; and Alcohol. Greater adherence to the Prudent DP was associated with favorable anthropometric outcomes. The Alcohol DP was associated with a favorable lipid profile. Associations between the Western DP and blood lipids differed by sex; with unfavorable impact observed only among males. Our findings add to the literature linking DPs in young adults with measurable adiposity and cardiometabolic outcomes; suggesting that improving nutrition among college students could reduce chronic disease risk.