Abstract: PO0798
Modifiable Risk Factors Associated with Death over the Long Interval for In-Center Hemodialysis Patients
Session Information
- Dialysis Care: Epidemiology and the Patient Experience
November 04, 2021 | Location: On-Demand, Virtual Only
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 701 Dialysis: Hemodialysis and Frequent Dialysis
Authors
- Harford, Antonia, Dialysis Clinic Inc, Nashville, Tennessee, United States
- Wilson, Jonathan A., Duke University, Durham, North Carolina, United States
- Paine, S., Dialysis Clinic Inc, Nashville, Tennessee, United States
- Ephraim, Patti, Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland, United States
- Pendergast, Jane F., Duke University, Durham, North Carolina, United States
- Majchrzak, Karen M., Dialysis Clinic Inc, Nashville, Tennessee, United States
- Miskulin, Dana, Tufts Medical Center, Boston, Massachusetts, United States
- Weiner, Daniel E., Tufts Medical Center, Boston, Massachusetts, United States
- Lacson, Eduardo K., Dialysis Clinic Inc, Nashville, Tennessee, United States
- Johnson, Doug, Dialysis Clinic Inc, Nashville, Tennessee, United States
- Johnson, Keith, Dialysis Clinic Inc, Nashville, Tennessee, United States
Background
Routine thrice weekly in-center hemodialysis (HD) is associated with increased risk of death around the long interval without dialysis. We evaluated this risk and explored modifiable risk factors associated with increased risk of death.
Methods
Prevalent Medicare beneficiaries (21,331) receiving maintenance thrice weekly in-center HD in facilities operated by a large not-for-profit dialysis provider from 1/1 2009 -12/31 2016 were included. We calculated daily death rates for an observation week (consisting of days HD-2, HD-1, HD1, HD+1, HD2, HD+2, and HD3) occurring after a week of usual outpatient care, defined by at least 1 outpatient treatment, with no inpatient hospital days, excused absences, or SNF days. HD1 represents the Monday or Tuesday following the long interval. A logistic regression model with patient random effects estimated the effect of patient characteristics, including age and treatment vintage, and treatment parameters including inter-dialytic weight gain (IDWG) >4.6% of estimated dry weight (kg), pre-HD systolic blood pressure (SBP) <120 mm Hg, and intradialytic hypotension, on the probability of dying over the long interval (HD-2, HD-1 and HD1) as compared to either dying on any other day in the week or survival.
Results
Among 2,019 deaths included in analyses, the highest death rates were observed on HD1 and HD3. Factors associated with increased odds of death over the long interval included older age, IDWG >4.6% of estimated dry weight, pre HD SBP <120 mm Hg prior to last treatment in the observation week, and skipped treatment in the prior week (table).
Conclusion
High inter-dialytic weight gain, low pre-HD SBP, and skipped treatments during the preceding week are potentially modifiable factors associated with increased risk of death over the long interval. These factors can help identify patients who will benefit from HD prescription modification.
| Variable | Comparison | Odds ratio for dying over long interval vs any other day or surviving (95% CI) |
| Inter-dialytic weight gain (kg) | ≧4.6% vs <4.6% | 1.767 (1.456, 2.145) |
| Age at start | One year change | 1.032 (1.025, 1.039) |
| Vintage | One year change | 1.055 (1.011, 1.100) |
| SBP absolute difference of 30 (mm Hg) pre to post treatment | Yes vs No | 0.886 (0.739, 1.061) |
| Pre-HD SBP <120 (mm Hg) | <120 vs ≧20 | 2.898 (2.468, 3.404) |
| Prior week treatment sessions (n) | <3 vs 3 | 3.079 (2.533, 3.742) |