Dietary patterns derived by Gaussian graphical models and metabolic profiles among overweight and obese individuals

The increasing prevalence of metabolic syndrome (MetS) and its associated risk factors highlights the critical need to understand dietary patterns that influence health outcomes, particularly among overweight and obese individuals. Identifying dietary networks provides valuable insights into the complex interactions between food groups within typical dietary patterns. This study aimed to derive dietary networks using Gaussian graphical models (GGM) and evaluate their associations with the risk of MetS components in a sample of the Iranian population.

This cross-sectional study involved 647 participants who were overweight or obese. The study included assessments of body composition and anthropometric measurements. Dietary fatty acid consumption was evaluated using a validated Food Frequency Questionnaire (FFQ) containing 168 items. Additionally, biochemical parameters, including serum total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), fasting serum glucose (FSG), and insulin levels, were measured using enzymatic methods. GGM was utilized to explore the networks of participants' dietary intake, and the association between these networks and risk factors related to MetS was assessed using logistic regression.

GGM analysis identified six major networks of dietary intake, where 28 food items were allocated into six dietary networks of vegetable, grain, fruit, snack, fish/dairy, and fat/oil dietary networks, with raw vegetables, grain, fresh fruit, snack, margarine, and red meat were central to the networks respectively. In the vegetable network, TC was significantly lower in the higher tertiles of this network, and HDL was higher in the third tertile compared with the first tertile in sex, age, and fully adjusted models. In the grain network, lower SBP, DBP, TG, LDL, and higher HDL were shown in the higher tertile (P < 0.05).

This study showed that vegetables and grains are associated with decreased risk of MetS components, including reduced blood pressure and cholesterol. Therefore, these results emphasize that dietary networks can be valuable for analyzing nutritional habits and consumption trends.