Associations Among Different Domains of Quality Among US Liver Transplant Programs

IMPORTANCE: US liver transplant programs have traditionally been evaluated on 1-year patient and graft survival. However, there is concern that a narrow focus on recipient outcomes may not incentivize programs to improve in other ways that would benefit patients with end-stage liver disease. OBJECTIVE: To determine the correlation among different potential domains of quality for adult liver transplant programs. DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study was conducted from 2014 to 2019 among adult liver transplant programs included in the United Network for Organ Sharing and Scientific Registry of Transplant Recipients program-specific reports. Liver transplant programs in the United States completing at least 10 liver transplants per year were included. Data were analyzed from March 2 to August 13, 2020. MAIN OUTCOMES AND MEASURES: The potential domains of quality examined included recipient outcomes (1-year graft and patient survival), aggressiveness (ie, marginal graft use, defined as the rate of use of donors with body mass index [calculated as weight in kilograms divided by height in meters squared] greater than 40, age older than 65 years, or deceased by cardiac death), and waiting list management (ie, waiting list mortality). The correlation among measures, aggregated at the center level, was evaluated using linear regression to control for mean Model for End Stage Liver Disease–Sodium score at organ allocation. The extent to which programs were able to achieve high quality across multiple domains was also evaluated. RESULTS: Among 114 transplant programs that performed a total of 44 554 transplants, the mean (SD) 1-year graft and patient survival was 90.3% (3.0%) with a total range of 75.9% to 96.6%. The mean (SD) waiting list mortality rate was 16.7 (6.1) deaths per 100 person-years, with a total range of 6.3 to 53.0 deaths per 100 person years. The mean (SD) marginal graft use rate was 15.8 (8.8) donors per 100 transplants, with a total range of 0 to 49.3 donors. There was no correlation between 1-year graft and patient survival and waiting list mortality (β = −0.053; P = .19) or marginal graft use (β = −0.007; P = .84) after correcting for mean allocation Model for End Stage Liver Disease–Sodium scores. There were 2 transplant programs (1.8%) that performed in the top quartile on all 3 measures, while 4 transplant programs (3.6%) performed in the bottom quartile on all 3 measures. CONCLUSIONS AND RELEVANCE: This cohort study found that among US liver transplant programs, there were no correlations among 1-year recipient outcomes, measures of program aggressiveness, or waiting list management. These findings suggest that a program’s performance in one domain may be independent and unrelated to its performance on others and that the understanding of factors contributing to these domains is incomplete.