NAFLD Diagnostic Nomogram for Community-Dwelling Elderly Using Routine Parameters

Objective: Given the increasing NAFLD burden in elderly populations and lack of dedicated prediction tools, we aimed to develop a nomogram using routine clinical parameters to predict NAFLD risk in community-dwelling elderly (≥65 years) in Shenzhen. Methods: This cross-sectional study enrolled 591 elderly residents who underwent health examinations between January-December 2024. NAFLD was diagnosed via abdominal ultrasound. The prediction model was constructed using LASSO and multivariate logistic regression, with internal validation through 1,000 bootstrap resamples. Performance was assessed via ROC curves and calibration plots. Results: Among 591 participants (mean age 71.65±6.55 years), NAFLD prevalence was 47.04%. The final model comprised eight predictors: BMI (OR=1.36, 95%CI:1.20-1.53), fasting glucose (OR=1.24, 95%CI:1.08-1.42), absence of hyperuricemia (OR=0.76, 95%CI:0.47-1.26), HDL cholesterol (OR=0.46, 95%CI:0.23-0.90), LDL cholesterol (OR=1.16, 95%CI:0.89-1.50), platelet count (OR=1.00, 95%CI:1.00-1.01), inverse of triglycerides (OR=0.27, 95%CI:0.13-0.56), and waist circumference (OR=1.01, 95%CI:0.97-1.05). The model demonstrated excellent discrimination (AUC=0.82) and calibration. Conclusion: We developed a well-performing NAFLD risk nomogram for the elderly, integrating eight routine clinical parameters. This tool may assist primary care providers in identifying high-risk individuals, optimizing screening strategies, and guiding interventions.