Background and Objectives
A simple screening tool is needed as an aid to optimize management of nonalcoholic fatty liver disease (NAFLD). The purpose of this study was to design a simple index that could be used to predict the presence of NAFLD.
Methods
A cross-sectional case–control study design was used for this study with 6,926 subjects who had received health assessments.
Results
Multivariate analysis indicated that body mass index (BMI), triglyceride (TG), alanine aminotransferase (ALT) to aspartate aminotransferase (AST) ratio (ALT/AST ratio), and hyperglycemia (HG) were independent risk factors for NAFLD. These variables were used to construct a fatty liver disease (FLD) index: FLD index = BMI + TG + 3 × (ALT/AST ratio) + 2 × HG (presence of HG, HG = 1; absence of HG, HG = 0). The FLD index had an area under the receiver-operating characteristic (AUROC) curve of 0.819 in the training set and 0.817 in the validation set. Values of the FLD index <28.0 or >37.0 excluded the possibility of NAFLD with sensitivity of 94.9 % or detected NAFLD with specificity of 96.0 %, respectively.
Conclusions
The FLD index is a simple, efficient NAFLD screening tool for the Chinese population that may be used to select people for further analysis and/or treatment, and/or for lifestyle modification.