This study used de-identified electronic health record (EHR) data drawn from 360 hospitals and 317,000 healthcare providers across all 50 U.S. states, covering 1999 to the present and representing roughly 20% of the U.S. population. The pooled records were normalized using a systematized nomenclature of clinical terms for liver disease or damage, then aggregated. Liver disease status was classified into six non-overlapping categories: alcoholic cirrhosis, non-alcoholic cirrhosis, chronic non-alcoholic liver disease, chronic Hepatitis C, Viral Hepatitis B, and alcoholic liver damage.