Mina Radmehr ¹ Reza Homayoun Far ^2* Abolghasem Djazayery ^3*

• ¹ Department of Nutrition, Science and Research Branch, Islamic Azad University of Medical Sciences, Tehran, Iran
• ² National Nutrition and Food Technology Research Institute, Shahid Beheshti University of Medical Sciences,, Tehran, Iran
• ³ Department of Community Nutrition, School of Nutritional Sciences and Dietetics, Tehran University of Medical Sciences, Tehran, Alborz, Iran

Background: Non-alcoholic fatty liver disease (NAFLD) is a widespread liver condition associated with diabetes, metabolic syndrome, and cardiovascular diseases, yet public awareness remains low. Early detection of risk factors is crucial, but liver biopsy, the diagnostic gold standard, is invasive and costly.Non-invasive anthropometric indices provide a safer alternative. This study examines these indices to identify the most reliable predictor of NAFLD in adults.In the present cross-sectional study, we used the Fasa Cohort Data, conducted on about 10,000 people, of whom 1,047 were diagnosed with NAFLD. NAFLD diagnosis in this study was confirmed by physicians based on medical history and ultrasonographic evaluations, ensuring accurate and reliable identification of cases. General, anthropometric, and dietary assessments were performed using interviews, tools, and valid questionnaires. Biochemical evaluation was also done. Waist-to-hip ratio (WHR), waist-to-height ratio )WHtR ( , Body mass index (BMI), a body shape index (ABSI), body roundness index (BRI), and visceral fat index (VAI) were also calculated using these measurements and formulas. This study used descriptive tests, binary logistic regression, and ROC curve analysis.In both crude and adjusted models, significant associations were found between WHR, WHtR, BMI, and VAI with NAFLD. ROC analysis revealed that WHtR and BMI were the most accurate predictors of NAFLD in both genders (WHtR: men AUC=0.750, women AUC=0.702; BMI: men AUC=0.754, women AUC=0.701). BRI showed significant accuracy, but WHR (men: AUC=0.727, women: AUC=0.640) and VAI (men: AUC=0.621, women: AUC=0.622) were less effective. ABSI demonstrated poor predictive power (men: AUC=0.530, women: AUC=0.505) and is not recommended for NAFLD prediction.Conclusions: Based on the findings, BMI and WHtR emerge as the most practical and accessible indicators for early screening of NAFLD in both men and women, while ABSI shows minor effectiveness in identifying the disease.