Background and Aims: Screening for liver fibrosis presents a clinical challenge. This study aimed to explore a useful alternative method for assessing fibrosis risk among US adults at risk of metabolic dysfunction-associated steatotic liver disease (MASLD). Methods: A liver stiffness score (LSS) model was proposed and tested using data from 3976 participants at possible risk of MASLD, obtained from the US National Health and Nutrition Examination Survey (NHANES). Results: The LSS model was developed using liver stiffness measurements, blood biochemistry, and body measurement data from 2414 NHANES participants at risk of MASLD, sampled between 2017 and 2020: LSS = exp(0.007035 × bodyweightkg - 0.1061 × raceblack1,0 + 0.183221 × diabetes1,0 + 0.008539 × ASTIU/L - 0.0018 × plateletcount1000cell/UL - 0.21011 × albuming/dL + 2.259087). The probability (P) of having fibrosis F3 + F4 is calculated as follows: P = 0.0091 × LSS2 - 0.0791 × LSS + 0.1933. The developed LSS model was tested on 1562 at-risk participants from the 2017-2018 cycle. The results showed that the LSS model achieved AUROC values of 0.79 and 0.78 for diagnosing cirrhosis (F4) and advanced fibrosis (F3 + F4) in the US population, respectively. It outperformed existing models such as NFS, FIB-4, SAFE, and FIB-3. For screening F3 + F4 fibrosis, the LSS model's top decile outperformed the NFS and FIB-4 models by 37.7% and 42.6%, respectively. Additionally, it showed superior performance compared to the waist circumference classification method by 29.5%. Conclusions: derived from an ethnically diverse population dataset, the LSS screening model, along with its probability equation, may offer clinicians a valuable alternative method for assessing the risk of liver fibrosis in the at-risk adult population.