Journal Home
Search for

Volume 50, Issue 5, Pages 774-781 (November 2007)


View previous. 19 of 33 View next.

The Relationship Between Dialysis Performance Measures: Adequacy and Anemia Management

Yonatan S. Spolter, BS1, Stephen L. Seliger, MD, MS12, Min Zhan, PhD2, Van Doren Hsu, PharmD3, Lori D. Walker, BS3, Jeffrey C. Fink, MD, MS12Corresponding Author Informationemail address

Received 15 January 2007; accepted 7 August 2007. published online 08 October 2007.

Refers to article:
Imperial or Empirical Measures of Dialysis Quality?
Klemens B. Meyer, Dana C. Miskulin
American Journal of Kidney Diseases
November 2007 (Vol. 50, Issue 5, Pages 691-695)
Full Text | Full-Text PDF (94 KB)
Background

Little is known about how urea reduction ratio (URR) and hemoglobin (Hb) level relate as clinical performance measures (CPMs) in dialysis facilities. This study examined the relationship between these CPMs as a reflection of underlying processes.

Study Design

Cross-sectional observational.

Setting & Participants

47,465 records from 18,003 patients dialyzed in 270 End-Stage Renal Disease Network 5 facilities during the final quarter of 2004.

Predictor & Outcome

Facility aggregate Hb level and URR, respectively; both expressed as (1) continuous variables and (2) dichotomous indicators based on established quality thresholds.

Measurements

Weighted regression analyses were used to determine the: (1) association between CPMs and (2) likelihood of missing the performance threshold for one if also missing the other benchmark.

Results

The association between facility URR and Hb level was minimal, with an adjusted mean increase (β) in URR of 0.91% ± 0.38% for each 1-g/dL increment in Hb level (P < 0.0001) and R2 of 0.02. The adjusted odds ratio of a facility failing to meet the URR quality threshold if missing the Hb level benchmark was 2.08 (P < 0.0001). The C statistic for the associated receiver operator characteristic curve was 0.64, with insignificant change when including Hb performance as a predictor of URR performance.

Limitations

Inability to incorporate historic performance or ascertain facility process traits that influence CPMs.

Conclusions

There was minimal association between facility URR and Hb level. The limited concordance in facility performance for these CPMs points to distinct processes that determine quality for each.

1 Department of Medicine, University of Maryland School of Medicine, Baltimore, MD

2 Department of Epidemiology and Preventive Medicine, University of Maryland School of Medicine, Baltimore, MD

3 Pharmaceutical Research Computing, University of Maryland School of Pharmacy, Baltimore, MD.

Corresponding Author InformationAddress correspondence to Jeffrey C. Fink, MD, MS, Rm N3W143, 22 S Greene St, University of Maryland Medical System, Baltimore, MD 21201.

 Originally published online as doi:10.1053/j.ajkd.2007.08.006 on October 4, 2007.

PII: S0272-6386(07)01142-0

doi:10.1053/j.ajkd.2007.08.006


View previous. 19 of 33 View next.