Abstract: TH-PO531
Does Vitamin D Level Explain the Racial Disparity in Albuminuria?
Session Information
- CKD: Health Services, Disparities, Prevention
November 02, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Chronic Kidney Disease (Non-Dialysis)
- 307 CKD: Health Services, Disparities, Prevention
Authors
- Lopes, Marcelo Barreto, Universidade Federal da Bahia , Salvador, BA, Brazil
- Bragg-Gresham, Jennifer L., University of Michigan, Ann Arbor, Michigan, United States
- Morgenstern, Hal, University of Michigan, Ann Arbor, Michigan, United States
- Burrows, Nilka Rios, Centers for Disease Control and Prevention, Atlanta, Georgia, United States
- Crews, Deidra C., Johns Hopkins University School of Medicine, Baltimore, Maryland, United States
- Powe, Neil R., Priscilla Chan and Mark Zuckerberg San Francisco General Hospital & University of California SF, San Francisco, California, United States
- Saran, Rajiv, University of Michigan, Ann Arbor, Michigan, United States
Background
Results from observational studies have shown that African Americans (AA) are more likely to have albuminuria than are whites and that 25-OH vitamin D3 (VitD) level is inversely associated with albuminuria. We hypothesized that VitD may partly explain the racial disparity in albuminuria risk by acting as a mediator in the causal pathway. Thus, the aim of this study was to estimate the direct and indirect (through VitD) effects of race on albuminuria in US adults.
Methods
Using cross-sectional data from 2007–2010 datasets from the National Health and Nutrition Examination Survey we applied the approach to mediation analysis of VanderWeele & Vansteelandt (2010) to estimate the direct and indirect effects of race on albuminuria prevalence, where VitD was treated as a continuous mediator. Estimated odds ratios (ORs with 95% CIs) were derived from: (1) a logistic model, where albuminuria prevalence was regressed on race, VitD, their product, and two potential confounders, age and sex; and (2) a linear model, where VitD was regressed on race, age, and sex. A secondary analysis adjusted for 6 additional risk factors for VitD or albuminuria. The total effect of race on albuminuria was estimated as the product of the direct and indirect effects.
Results
AA individuals had a higher prevalence of VitD<50nmol/L (72.4% vs. 20.5%. p<0.001) and albuminuria (11.2% vs. 8.6%. p=0.025) compared to Whites. The OR for the direct effect of race on albuminuria was 1.42 (95% CI: 1.10, 1.94), and the OR for the indirect effect was 1.29 (1.14, 1.46). The OR for the total race effect was 1.83, and the proportion of the race effect mediated by VitD was 42%. Additional adjustment for 6 other potential confounders yielded weaker direct and indirect effects but increased the proportion mediated to 51%.
Conclusion
The effect of VitD may explain nearly half the racial disparity in albuminuria prevalence in the United States; however, causal inference is limited by the cross-sectional design and possible residual confounding. Future research may address these limitations and consider other possible mediators of the race effect.
Funding
- Other U.S. Government Support