Abstract: PO1250
Geographic Variation of Home Dialysis Utilization in the United States
Session Information
- Peritoneal Dialysis - 1
October 22, 2020 | Location: On-Demand
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 703 Dialysis: Peritoneal Dialysis
Authors
- Weinhandl, Eric D., Chronic Disease Research Group, Minneapolis, Minnesota, United States
- Gilbertson, David T., Chronic Disease Research Group, Minneapolis, Minnesota, United States
- Wetmore, James B., Chronic Disease Research Group, Minneapolis, Minnesota, United States
- Johansen, Kirsten L., Chronic Disease Research Group, Minneapolis, Minnesota, United States
Background
Increasing home dialysis utilization is an aim of the Advancing American Kidney Health Initiative. We estimated geographic variation in home dialysis utilization in a contemporary population of end stage kidney disease (ESKD) patients. We also assessed the extent to which race and payers—possible systemic barriers to home dialysis—account for this variation.
Methods
Using USRDS Standard Analysis Files, we identified all prevalent ESKD patients on December 31, 2017 and ascertained the dialysis modality—in-facility hemodialysis, home hemodialysis (HHD), or peritoneal dialysis (PD)—of each patient on that date. We categorized patients into 306 Hospital Referral Regions (HRRs), according to ZIP code of the dialysis facility. We estimated the standardized home dialysis ratio (SHDR) in each HRR, with expected home dialysis utilization as a logistic regression of age, sex, primary cause of ESKD, and ESKD duration. Subsequently, we added race and payer to the regression.
Results
The cohort comprised 513,669 patients. Home dialysis utilization was 12.0% (1.8% HHD, 10.2% PD). Among HRRs, 5th and 95th percentiles of observed utilization were 5.3% and 23.2%, respectively, whereas 5th and 95th percentiles of SHDR were 0.43 and 1.82, as displayed. There were 87 HRRs (28%) with SHDR significantly <1.0 (P < 0.05) and 116 (38%) with SHDR significantly >1.0. Of the 10 HRRs with largest patient counts, seven—Los Angeles, Houston, Manhattan, Dallas, East Long Island, Philadelphia, and San Antonio—had SHDR significantly <1.0. The addition of race and payer improved the discrimination of logistic regression, with black race and concurrent Medicaid enrollment as negative predictors of home dialysis utilization. However, the distribution of SHDR did not greatly compress. There were 29 HRRs (9%) with SHDR that was revised from significant to non-significant and 22 (7%) with SHDR that was revised from non-significant to significant.
Conclusion
Large geographic variation in home dialysis utilization exists. Race and payer are associated with utilization, but adjustment for these factors does not alter variation in SHDR.
Standardized Home Dialysis Ratio
| Percentiles | |||||||
| 1st | 5th | 25th | 50th | 75th | 95th | 99th | |
| SHDR | 0.26 | 0.43 | 0.80 | 1.06 | 1.33 | 1.82 | 2.40 |
| SHDR (revised) | 0.24 | 0.42 | 0.79 | 1.05 | 1.30 | 1.73 | 2.29 |
In the revised SHDR, expected home dialysis utilization also reflects race and payer.
Funding
- NIDDK Support