Empirically-Derived Dietary Patterns in Relation to Non-Alcoholic Fatty Liver Diseases Among Adult Participants in Amol, Northern Iran: A Structural Equation Modeling Approach

Dietary modifications remain the mainstay in managing nonalcoholic fatty liver disease (NAFLD). Published data on the effect of overall dietary patterns on NAFLD is scarce. The present study aims to extract the dietary patterns and investigate their association to NAFLD by gender, using structural equation modeling, among adult participants in Amol, northern Iran. In this cross-sectional study, data from 3,149 participants in the Amol cohort study (55.3% men, n = 1,741) were analyzed. Usual dietary intake was assessed by a validated 168-items semiquantitative food frequency questionnaire. We classified major dietary patterns by explanatory factor analysis (EFA) and confirmatory factor analysis (CFA). NAFLD diagnosis was based on ultrasound scanning, including increased hepatic echogenicity, abnormal appearance of hepatic arteries, and diaphragm in the absence of excessive alcohol consumption. Multivariable logistic regression and structural equation modeling (SEM) were used to explore the relationship between dietary patterns and NAFLD. Three distinct dietary patterns, including western, healthy, and traditional/mixed dietary patterns, were identified. Adult male who adhere to the western dietary pattern were more affected with NAFLD risk [Q1, Q2, Q3, Q4, odds ratio (OR) = 1, 1.16, 1.34, 1.39; 95% confidence interval (CI) = 0.83–1.61, 0.96–1.85, 0.98–1.96, p(trend) = 0.04, respectively]. A full mediating effect of healthy dietary pattern, western dietary pattern, and traditional dietary pattern via dietary acid load (DAL) proxy (of dietary patterns to DAL: βstd = −0.35, p < 0.006, βstd = 0.15, p = 0.009, and βstd = 0.08, p = 0.001, respectively), on NAFLD was found through mediation analysis using SEM. A western dietary pattern comprising frequent intake of salty and sweet snacks, soft drinks, refined grains, processed meats, cooked and fried potatoes, eggs, and coffee was associated with a higher odds of NAFLD in an Iranian male population. Additionally, our findings might provide a mechanistic explanation for the association between dietary patterns and NAFLD via DAL proxy. However, further prospective studies, including assessing acid-base biomarkers, are needed.