Anemia and new-onset congestive heart failure in the general medicare population

Peter E. Sandgren, Anne M. Murray, Charles A. Herzog, Craig A. Solid, David T. Gilbertson, Allan J. Collins, Robert N. Foley

Research output: Contribution to journalArticlepeer-review

27 Scopus citations


Background: Anemia is associated with disease severity and prognosis for patients with congestive heart failure (CHF). It is unknown whether anemia is associated with the development of new-onset CHF in the elderly. Methods and Results: This retrospective longitudinal cohort study used the Medicare 5% database. In the incident analysis, the study sample comprised subjects without CHF in 1999 (n = 1,063,495); the main exposure variable evaluated was the presence or absence of anemia in 1999; and the primary study outcome was the occurrence of new-onset CHF. The prevalence of chronic anemia and CHF in 1999 was 5.0% and 9.9%, respectively. The incidence of new-onset CHF in 2000 in those with and those without anemia in 1999 was 12.3% and 5.9%, respectively, corresponding to an adjusted hazard ratio of 1.29 (P <. 001), a value intermediate between the hazard ratios of 1.13 (P <. 001), and 1.76 (P <. 001) associated with hypertension and atherosclerotic heart disease, respectively. Anemia also was associated with death in the year after new-onset CHF (adjusted hazard ratio, 1.28; P <. 001). Conclusion: Anemia in the Medicare population is associated with the diagnosis and prognosis of new-onset CHF.

Original languageEnglish (US)
Pages (from-to)99-105
Number of pages7
JournalJournal of cardiac failure
Issue number2
StatePublished - Mar 2005

Bibliographical note

Funding Information:
Supported by an unrestricted research grant from Amgen Inc., Thousand Oaks, California, and by the Minneapolis Medical Research Foundation, Minneapolis, Minnesota.


  • Death
  • Prognosis
  • Retrospective studies


Dive into the research topics of 'Anemia and new-onset congestive heart failure in the general medicare population'. Together they form a unique fingerprint.

Cite this