Neighborhood Disadvantage and Access to Liver Transplant Referral for Severe Alcohol-Associated Hepatitis

Abstract

Importance: Severe alcohol-associated hepatitis (sAH) is a leading indication for liver transplant (LT). However, access to LT begins with referral, a step that may not have been previously characterized using prospective multicenter data with detailed measures of social determinants of health. Objective: To examine the association of clinical severity and neighborhood disadvantage with referral, wait-listing, transplant, and short-term mortality in patients hospitalized with sAH. Design, setting, and participants: This prospective multicenter cohort study was a secondary analysis of the Alcohol-Associated Hepatitis Network observational cohort. It included patients hospitalized with sAH (Model for End-Stage Liver Disease [MELD] >20) across 5 US transplant centers from May 6, 2019, to November 8, 2023. Clinical, demographic, and social determinants of health data, including the Area Deprivation Index (ADI), were collected prospectively. Exposures: Clinical severity (MELD score) and neighborhood-level disadvantage (ADI). Main outcomes and measures: The primary outcomes were referral for LT evaluation, wait-listing, and receipt of LT, and the secondary outcome was 180-day mortality. Logistic regression and generalized additive models were used to evaluate independent and interactive associations of MELD and ADI with outcomes. Results: The cohort included 325 patients (mean [SD] age, 44.8 [10.2] years; 197 males [60.6%]). The mean (SD) MELD score was 29.2 (7.6), and the mean (SD) ADI was 56.2 (24.4). Only 120 patients (36.9%) were referred for LT. In multivariable analysis, higher MELD scores were associated with lower referral odds (odds ratio, 1.13 [95% CI, 1.07-1.18]; P < .001). ADI was not associated with referral odds (odds ratio, 0.99 [95% CI, 0.97-1.00]; P = .06). Generalized additive model analyses demonstrated significant MELD × ADI interactions for referral (P for interaction = .01), wait-listing (P for interaction = .004), and mortality (P for interaction = .01). At a MELD score 20 to 30, referral probability was 40% to 60% among patients with an ADI less than 30 vs 20% among those with an ADI 30 or more. Among patients with a MELD score 20 to 30, mortality exceeded 20% in those with an ADI 60 or more compared with 10% to 20% in those with an ADI less than 20. Conclusions and relevance: In this multicenter cohort study of severe alcohol-associated hepatitis, neighborhood disadvantage was associated with referral at intermediate MELD scores. These findings suggest that referral is a critical leverage point for interventions and that outreach and navigation strategies informed by social context may improve access to LT.