S2538 Validation of NAFLD ICD Codes to Identify MASLD in a High-Risk Hispanic Population

Introduction: Metabolic dysfunction-associated steatotic liver disease (MASLD), a common cause of chronic liver disease, can progress to cirrhosis and hepatocellular carcinoma. Through consensus, nonalcoholic fatty liver disease (NAFLD) was renamed MASLD. Given distinct case definitions, overlap between these diagnoses is variable. The global prevalence of MASLD is 30%, however risk is highest in Hispanics. This study aims to characterize the validity of historical ICD codes for NAFLD for MASLD diagnosis within a high-risk Hispanic patient population. Methods: Using research informatics, we identified Hispanic adults with International Classification of Diseases, Ninth Revision (ICD-9) codes 571.8/571.9 and ICD-10 codes K76.0 and K75.81 without codes for alcohol use disorder or other chronic liver diseases seen from July 2024-December 2024. We reviewed the electronic medical records (EMR) of the subset of patients seen within the Gastroenterology or Hepatology clinics (GI/Hep). We characterize the proportion of patients with MASLD, metabolic dysfunction and alcohol-related liver disease (MetALD), or other liver diseases. Results: There were 7,183 patients with relevant ICD-9/10 codes for NAFLD, of whom 4,688 were of Hispanic, 65.3%. Only 33.2%, n = 1,558, were seen by GI/Hep. Within that group, the median age was 58 years and 62% were women. According to race listed in the EMR, the sample was 95.3% White, 2% Multiracial, 1.8% Black, and 0.9% Other. Our Hispanic population is diverse, representing 35 countries of origin; 21.4% were US-born and 53% have limited English proficiency. Of these 1,558 patients identified using codes for NAFLD, 1,511 (97%) had fatty liver present. Only 1,246 (80%) met diagnostic criteria for MASLD, 98 (6.3%) met criteria for MetALD, and 16 (1%) met criteria for ALD. There were 198 patients who did not meet criteria for MASLD because they had other concurrent liver diseases (151, 10%) or no evidence of steatosis (47, 3%). Conclusion: Previous reports have suggested high concordance of NAFLD and MASLD, ranging from 95%-99%. We found high rates of steatosis, but expected higher concordance between NAFLD and MASLD, given that these patients were seen in subspecialty clinics and the Hispanic population has higher risk. Our results highlight the risk of diagnostic misclassification when using administrative data alone for case-finding and the need for prospective validation of MASLD diagnostic codes using novel diagnostic criteria.