Structural Bias in Renal Allograft Expected Outcomes Calculations in the Southern US: A County-Level Analysis Using the County Deprivation Index.

Abstract

Socioeconomic deprivation and limited healthcare access contribute to persistent regional disparities in transplant outcomes across the US Traditional risk-adjustment models, such as those used by the Scientific Registry of Transplant Recipients (SRTR), account for recipient and donor clinical factors but exclude structural determinants of health. Whether area-level socioeconomic context explains residual variation in graft failure beyond clinical risk remains unclear, particularly in the Southern US. A population-based ecological study linked adult deceased-donor kidney transplant recipients from the US Renal Data System (2016 to 2019) with county-level socioeconomic measures, including the County Deprivation Index (CDI). Expected 3-year graft failure risks were estimated using SRTR model coefficients applied to donor and recipient characteristics, then aggregated by county. Weighted Poisson regression models, using expected failures as an offset, evaluated associations between CDI and observed-to-expected (O/E) graft failure ratios across 12 southern states. Restricted cubic spline models demonstrated a nonlinear CDI-graft failure relationship (p = 0.01). O/E ratio increased sharply with deprivation and plateaued beyond CDI ≈ 2.5, suggesting a ceiling effect in disadvantaged counties. In models truncating CDI at 2.5, each 1-unit increase in CDI corresponded to a 9% higher O/E graft failure rate (rate ratio 1.09; 95% CI 1.09 to 1.10; p < 0.001). Deprivation effects were strongest among counties with the lowest predicted clinical risk. County-level socioeconomic deprivation is independently associated with excess graft failure beyond SRTR-predicted expectations. Incorporating contextual measures such as the CDI into national performance-reporting frameworks may improve fairness and accuracy in posttransplant outcome evaluation.