OBJECTIVES: Prevalence estimates of MASH (formerly NASH, non-alcoholic steatohepatitis) range from 115-506 cases per 100,000 adults. MASH is considered common among people ≥65 years of age but the exact prevalence is unknown. This analysis characterized the epidemiology of MASH among US older adults.

METHODS: Medicare 100% fee-for-service claims (MCR; 2016-2021), Optum’s de-identified Clinformatics® Data Mart Database (CDM) and Veradigm’s electronic health record (EHR)- linked to Komodo claims (both 2015-2022). NASH was defined by ≥1 inpatient claim with a primary or secondary diagnosis of K75.81 (NASH) or ≥2 outpatient claims. Within Veradigm, NASH was additionally defined by ≥1 EHR record with a NASH diagnosis. Data collection was restricted to those ≥65 years. Cases with individuals featuring ≥6 months of baseline data before first NASH diagnosis were considered incident. Demographics, incidence, and prevalence per 100,000 were summarized. A range of period prevalence estimates was estimated using: denominators based on those with 6- and 12-months follow-up; prevalence in the midpoint year; and average numbers of older adults across study period. Annual prevalence was scaled to approximate the frequency in the 2022 US population ≥65 years.

RESULTS: There were 51,343 (MCR),14,982 (CDM), and 144,172 (Veradigm) adults ≥65 years of age with NASH identified. Over follow-up, annual incidence was 11.6-26.5 (MCR), 26.9-49.4 (CDM), and 67.0-74.5 (Veradigm) per 100,000. Prevalence increased annually from 2016 to 2020, ranging from 49.2-91.4 in MCR, 49.0-176.6 in CDM, and 92.5-158.6 in Veradigm. Period prevalence was 169.5-199.3 (MCR), 143.3-286.5 (CDM), and 357.8-594.8 (Veradigm). In 2022, the extrapolated US older adult population with NASH was 47,701 (MCR), 81,007 (CDM), and 99,115 (Veradigm).

CONCLUSIONS: Findings demonstrate substantial and increasing prevalence of MASH among older adults. As the definition of NASH used within CDM and MCR was more specific vs. the broader definition within Veradigm, estimates from these databases provide lower and upper bounds.