Development and Validation of a Nomogram for Prediction of the Risk of MAFLD in an Overweight and Obese Population

BACKGROUND AND AIMS: Metabolic associated fatty liver disease (MAFLD) is a serious condition, and a simple method is needed for practitioners to identify patients with the disease and have a high risk of disease progression. METHODS: We developed and validated a nomogram for fatty liver disease and reclassified the risk factors for MAFLD. The development cohort had 335 patients who received bioelectrical impedance analysis and liver ultrasound attenuation measurements at Shenzhen People’s Hospital between September 2020 and June 2021. The validation cohort had 200 patients from other hospitals who received the same evaluation. A random forest procedure and binary logistic analysis were used to screen for risk factors, establish a fatty liver disease predictive model, and forecast the risk of MAFLD. The performance of the nomogram was evaluated by measurement of discrimination, calibration, and clinical usefulness. RESULTS: The nomogram provided good predictions in a model that included body mass index (BMI) and waist circumference. The areas under the curve of the nomogram were 0.793 in the development cohort and 0.774 in the validation cohort. The nomogram performed well for calibration, category-free net reclassification improvement, and integrated discrimination improvement. Decision curve analysis indicated the nomogram performed better than BMI for predicting net outcome. CONCLUSIONS: The nomogram was an effective screening tool for fatty liver disease, and for those overweight individuals, may help physicians make appropriate decisions regarding treatment of MAFLD.