Background: Racially minoritized populations in the United States (US), notably African American (AA) and American Indian/Alaska Native (AI/AN), experience disproportionately higher rates of chronic kidney disease (CKD), diabetes, and hypertension compared to their White peers but are understudied. This real-world cohort study examines the standards of CKD care provided to these groups in two US health systems.
Methods: Using electronic health record data from the Center for Kidney Disease Research, Education, and Hope (CURE-CKD) Registry (N = 381,011) collected between 2015 and 2020, adjusted binary logistic regression models were used to identify predictors of two CKD care outcomes: 1) prescriptions for CKD-related guideline-directed medical therapy (CKD-GDMT) in the form of angiotensin converting enzyme inhibitors or angiotensin receptor blockers and 2) testing for urine albumin-creatinine/urine protein-creatinine ratio (UACR/UPCR) among adult patients of AA and AI/AN race compared to the reference group (White, non-Hispanic).
Results: AA (62 ± 17 years) and AI/AN (57 ± 18 years) patients with CKD were younger compared to the White, non-Hispanic reference group (68 ± 17 years). Diabetes and hypertension were the most important predictors for increased odds of CKD-GDMT and UACR/UPCR testing. Prevalence of CKD-GDMT was only 46%, 40% and 38% in AA, White, and AI/AN patients, respectively. AA patients were more likely to receive CKD-GDMT prescriptions (OR = 1.20, 95% CI: 1.17-1.23, p < 0.001) and UACR/UPCR testing (OR = 1.34, 95% CI: 1.29-1.38, p < 0.001) compared to White patients. AI/AN were also more likely to receive GDMT (OR = 1.07, 95% CI: 1.00-1.15, p = 0.046) compared to White patients but had lower odds of UACR/UPCR testing (OR = 0.73, 95% CI: 0.67-0.81, p < 0.001). However, the frequency or prescribing of CKD-GDMT and UACR/UPCR testing were suboptimal across all examined racial identity groups. Exploratory machine learning approaches, including logistic regression, lasso regression, and random forest found similar findings.
Conclusions: While there were modest racial differences in the prescription of CKD-GDMT and frequency of UACR/UPCR testing, rates were lower than expected in this high-risk population. Our findings suggest the disproportionate burden of CKD on AA and AI/AN individuals is not solely attributable to the current standards of care delivery. The relatively higher rates of CKD-GDMT for AA patients may be due to clinician recognition of their increased risk for progressing to kidney failure.
Clinical trial number: Not applicable.
Keywords: Albuminuria; Angiotensin converting enzyme inhibitors; Angiotensin receptor blockers; Guideline-directed medical therapies; Machine learning; Racial health disparities.
© 2025. The Author(s).