Abstract: SA-OR038

Scheduled versus Emergency Dialysis in ESRD Saves Lives and Lowers Utilization: A Quasi-Randomized Study

Session Information

Category: Dialysis

  • 607 Dialysis: Epidemiology, Outcomes, Clinical Trials - Non-Cardiovascular

Authors

  • Nguyen, Oanh Kieu, UT Southwestern Medical Center, Dallas, Texas, United States
  • Vazquez, Miguel A., UT Southwestern Medical Center, Dallas, Texas, United States
  • Charles, Lakeesha, Parkland Health & Hospital System, Dallas, Texas, United States
  • Berger, Joseph Rossi, UT Southwestern Medical Center, Dallas, Texas, United States
  • Quinones, Henry, UT Southwestern Medical Center, Dallas, Texas, United States
  • Fuquay, Richard C., Dallas Nephrology Associates, Dallas, Texas, United States
  • Halm, Ethan, UT Southwestern Medical Center, Dallas, Texas, United States
  • Makam, Anil N, UT Southwestern Medical Center, Dallas, Texas, United States
Background

In many states across the U.S., individuals with end-stage renal disease (ESRD) lacking federal funding for scheduled dialysis instead receive intermittent emergency dialysis only for life-threatening indications. The effects on health outcomes and utilization compared to scheduled dialysis are unknown. We sought to compare these strategies through a natural experiment among individuals with ESRD on emergency dialysis who were newly eligible and applied for private insurance coverage for scheduled dialysis, with nearly half receiving coverage in a quasi-randomized fashion.

Methods

Retrospective cohort study of 193 adults on emergency dialysis in Dallas, Texas who applied for private insurance in February 2015. Patient characteristics and outcomes were ascertained using medical record and all-payer regional claims data. Overall, 112 were enrolled in scheduled dialysis; 81 were declined for non-patient-related reasons (i.e., their dialysis center declined to participate) and remained on emergency dialysis (control). We compared emergency department (ED) visits and hospitalizations in a 6-month baseline and a 12-month follow-up period after enrollment with a 1-month washout, using intention-to-treat negative binomial difference-in-differences (DiD) regression analyses.

Results

At baseline, the scheduled group was younger (45 vs. 52 yrs., p<0.001), had more frequent dialysis (1.0 vs 0.6 sessions/week, p=0.04), more ED visits (7.6 vs. 4.3 median visits/month, p<0.001), and similar hospitalizations (median 2.5 vs. 3.2 per 6 months, NS) vs. controls. After enrollment, the scheduled group had fewer deaths (6% vs. 16%, p<0.01). In adjusted analyses, compared to baseline, the scheduled group had a larger net decrease in ED visits (-6.7 vs. -0.1 visits/month, p<0.001; DiD of 6.6 fewer visits/month, 95% CI 5.5-7.7) and similar net decrease in hospitalizations (-1.7 vs. -1.4 per 6 months, p=0.46) during follow-up vs. controls.

Conclusion

In this quasi-randomized controlled study, individuals enrolled in a scheduled dialysis program had far lower rates of death and ED utilization in the year after enrollment compared to those remaining on emergency dialysis, despite being sicker at baseline. Universal scheduled dialysis improves health outcomes and may also be more cost-effective.

Funding

  • Other NIH Support