Hemodialysis treatment center early mortality rates for incident hemodialysis patients are associated with the quality of care prior to starting but not following onset of dialysis

Abstract

Background: We examined the independent contribution of pre-ESRD (end-stage renal disease) care and care after starting hemodialysis (post-HD) with facility-specific mortality among incident patients.

Methods: We studied 6,217 incident patients treated at 311 dialysis facilities. A pre-ESRD care score was assessed as the sum of quality measures met on the Centers for Medicare and Medicaid Services Form 2728, including predialysis nephrology and dietary care, having a fistula, hemoglobin and serum albumin. A post-HD care score was evaluated by the sum of quality targets attained, including HD adequacy, anemia, serum albumin and hemoglobin measured on an annual quality survey. A fifth post-HD care measure was having obtained an influenza vaccination during the current year.

Results: Individual patient mortality was associated with both pre-ESRD (p < 0.001) and post-HD (p < 0.001) care scores. Linear regression models including both pre-ESRD and post-HD care scores showed that a 1-point increase in the pre-ESRD care score resulted in a 0.30 (95% CI: -0.47, -0.12) decreased facility standardized mortality ratio; no association for post-HD care score was noted (-0.11; 95% CI: -0.26, 0.04).

Conclusion: Pre-ESRD and post-HD care are both strongly associated with individual patient mortality. In contrast, only pre-ESRD care is associated with facility mortality, suggesting that early mortality reflects differences in pre-ESRD care in the community.