Dietary and Lifestyle Predictors of Obesity: A Structural Equation Model Approach

Background Obesity and overweight, driven by complex interactions between socioeconomic, behavioral, and dietary factors, remain critical global health challenges. This study employs structural equation modeling to evaluate direct and indirect effects of these factors, mediated by dietary indices(Dietary Inflammatory Index and Healthy Eating Index), on obesity/overweight in western Iran. Methods A cross-sectional analysis of 3,169 adults from the Dehgolan Prospective Cohort Study(DehPCS) in western Iran was conducted. Data included anthropometric measures, socioeconomic status, lifestyle factors, chronic diseases, and dietary indices(HEI and DII). Structural equation modeling evaluated pathways linking latent variables and mediators to obesity/overweight. Model fit was assessed using CFI, RMSEA, and χ² statistics. Results The model explained 69% of obesity/overweight variance. Socioeconomic status directly reduced obesity risk (β=−0.21,p < 0.001), while chronic diseases increased it (β = 0.42,p = 0.002).Personal habits had a strong inverse association (β=−1.85,p < 0.001). DII elevated obesity likelihood (β = 0.65, p < 0.001), whereas HEI-improved diets reduced it (β=−1.14,p < 0.001). Socioeconomic status indirectly worsened outcomes via unhealthy dietary habits (DII: β = 0.01; HEI: β=−0.87), while chronic diseases indirectly lowered risk through improved HEI(β = 0.11,p = 0.03). Conclusion Obesity is mediated by dietary quality and inflammation, with socioeconomic disparities and chronic conditions amplifying risk. Public health strategies must prioritize anti-inflammatory diets and equitable access to nutritious foods to disrupt these pathways.