*Purpose: Recently, the use of socially constructed “race” in kidney disease research has received national attention partly because of the research bias created when race is defined as having biologic value. Like imprecise definitions, measurement error also leads to research bias. However, the effect of race measurement error in research bias in kidney transplantation research has not been examined.

*Methods: We used national, transplant registry data and data from a prospective, multi-center cohort of adult kidney transplant (KT) candidates and recipients that collects self-reported race from 12/2008-2/2020 to assess agreement in racial classification for same KT candidates and recipients and examine changes in the agreement over time for each racial group.

*Results: While concordance was almost perfect for the overall population (Kappa statistic (95% CI) = 0.92 (0.91, 0.93), it varied by race with almost perfect concordance being observed among White and Black populations, and moderate to weak among Hispanic/Latino and Native Hawaiian/Other Pacific Islander populations (Kappa statistic (95% CI) for White = 0.95 (0.95, 0.96), Black = 0.97 (0.96, 0.98), Asian = 0.88 (0.84, 0.92), Native Hawaiian/Other Pacific Islander = 0.40 (0.01, 0.78), Hispanic/Latino = 0.66 (0.57, 0.74), and Other = 0.08 (-0.01, 0.17)). For the White population, concordance decline linearly over time (Correlation coefficient between agreement and time = -0.43, p<0.001).

*Conclusions: Our results support the dynamic and inherently imperfect social construction and assignment of “race” to KT candidates and recipients. Race misclassification has likely biased racial disparities research in organ transplantation, especially for KT candidates and recipients who are not perceived to be or report being White or Black.

« Back to 2022 American Transplant Congress