Abstract
Despite national guidelines, medical practices and kidney transplant waiting list registration policies may differ from one dialysis/transplant unit to another. Benefit risk assessment variations, especially for elderly patients, have also been described. The aim of this study was to identify sources of variation in early kidney transplant waiting list registration in France. Among 16,842 incident patients during the period 2016-2017, 4,386 were registered on the kidney transplant waiting list at the start of, or during the first year after starting, dialysis (26%). We developed various log-linear mixed effect regression models on 3 levels: patients, dialysis networks and transplant centers. Variability was expressed as variance from the random intercepts (+/- standard error). Although patient characteristics have an important impact on the likelihood of registration, the overall magnitude of variability in registration was low and shared by dialysis networks and transplant centers. Between-transplant center variability (0.23 +/- 0.08) was 1.8 higher than between-dialysis network variability (0.13+/- 0.004). Older age was associated with a lower probability of registration and greater variability between networks (0.04 , 0.20, 0.93 in the 18-64, 65-74 and 75-84 age groups) . Targeted interventions should focus on elderly patients and/or certain regions with greater variability in waiting list access.