Abstract: FR-PO520
Lower Utilization of Home Dialysis in Non-White and Dual-Eligible Dialysis Patients
Session Information
- Peritoneal Dialysis: Modality, Catheter, Infections
November 08, 2019 | Location: Exhibit Hall, Walter E. Washington Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 703 Dialysis: Peritoneal Dialysis
Authors
- Ray, Debabrata, Fresenius Medical Care North America, Waltham, Massachusetts, United States
- Weinhandl, Eric D., Fresenius Medical Care North America, Waltham, Massachusetts, United States
Background
Past studies have associated non-white race with lower utilization of home dialysis. We aimed to assess the joint influence of race and dual eligibility (i.e., concurrent enrollment in Medicare and Medicaid), a surrogate for poverty, on home dialysis utilization in each state.
Methods
Using Medicare Limited Data Sets, we identified all patients with Medicare Part B claims documenting outpatient dialysis from January 2014 to December 2017. For each calendar week (Monday to Sunday), we identified patients with at least one day of either home hemodialysis or peritoneal dialysis treatment. Within each of the 50 states, we modeled utilization of home dialysis with logistic regression, using a generalized estimating equation. Risk factors comprised non-white race, dual eligibility (DE), the interaction of race and DE, age, sex, and calendar year. We applied a Bonferroni correction to the set of tests of race, DE, and the interaction thereof (α = 0.05/150).
Results
The cohort comprised 633,288 patients and 61,522,192 patient-weeks. Overall, 1.8% of patient-weeks were marked by home hemodialysis and 8.9% were marked by peritoneal dialysis.
Relative to white patients without DE, state-specific adjusted odds ratios (AORs) of home dialysis in non-white patents without DE were centered at 0.58; among 34 states in which the effect of race was significant, AORs ranged from 0.28 to 0.76. State-specific odds ratios of home dialysis in white patients with DE were centered at 0.42; among 43 states in which the effect of DE was significant, AORs ranged from 0.25 to 0.78.
Relative to white patients without DE, state-specific AORs of home dialysis in non-white patients with DE were centered at 0.25. However, interactions of race and DE were significant in only 2 states. In 46 states, non-white patients with DE were less likely to undergo home dialysis than both non-white patients without DE and white patients with DE.
Conclusion
Both non-white race and DE are associated with lower likelihood of home dialysis utilization in most states. The effects are generally log-additive, thus placing non-white patients with DE at profoundly lower likelihood of home dialysis utilization. Substantial growth in home dialysis will necessitate identifying and addressing the reasons for these disparities.