Time to target haemoglobin concentration (11 g/dl)--risk of hospitalization and mortality among incident dialysis patients

Nephrol Dial Transplant. 2007 Aug;22(8):2247-55. doi: 10.1093/ndt/gfm374. Epub 2007 Jun 27.

Abstract

Background: Haemoglobin levels <11 g/dl are associated with increased costs, morbidity and mortality. We aimed to determine if time required to reach 11 g/dl was associated with increased risk of hospitalization or death among incident dialysis patients.

Methods: We studied 29 131 patients initiating dialysis in 2002 and surviving >or=9 months. Demographic, comorbid and health care use data were extracted from Medicare claims from months 4-9 post-dialysis initiation. Logistic regression was used to calculate a propensity score for odds of longer than mean time to target. Proportional hazard models were used to assess effects of longer time on hospitalization and death. Other models were stratified for quartile of propensity score.

Results: Mean time to target haemoglobin was 1.3 months and 36% of participants required longer. These were more likely to be younger and minority, to use a dialysis catheter, and to have more comorbidity and hospitalization days during the entry period. Longer time to target was associated with increased risk for hospitalization (hazards ratio 1.15; 95% confidence interval 1.12-1.19) and mortality (1.26; 1.20-1.33) in the following year. Associations did not change when stratified by quartile of propensity score.

Conclusions: Longer time required to reach the target haemoglobin level was associated with significantly higher risk of hospitalization and mortality. Whether observed associations resulted from residual confounding by more severe illness remains unclear. Future trials should determine if rapidity of haemoglobin correction influences outcomes.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Chemistry, Clinical / methods*
  • Child
  • Child, Preschool
  • Female
  • Hemoglobins / analysis*
  • Hospitalization / statistics & numerical data*
  • Humans
  • Infant
  • Kidney Diseases / mortality*
  • Kidney Diseases / therapy*
  • Male
  • Middle Aged
  • Proportional Hazards Models
  • Regression Analysis
  • Renal Dialysis / adverse effects*
  • Treatment Outcome

Substances

  • Hemoglobins