OBJECTIVES: This study evaluated recent erythropoiesis-stimulating agent (ESA) utilization and costs from 2 large health insurance claims databases in chronic kidney disease (CKD) patients not on dialysis.  METHODS: An analysis of recent medical claims from the Ingenix IMPACT (January 2006-March 2009) and Medicare 5% (January 2005-December 2007) databases was conducted. Patients ≥18 years, newly initiated on epoetin alfa (EPO) or darbepoetin alfa (DARB), with ≥ 1 claim for CKD and ≥2 ESA claims were included. Patients diagnosed with cancer, receiving chemotherapy or dialysis, or receiving both agents were excluded. Mean cumulative ESA dose was used to calculate drug cost (using October 2009 wholesale acquisition cost) and dose ratio (Units EPO: mcg DARB). Multivariate analysis was also conducted to assess adjusted cost differences between the two agents.  RESULTS: A total of 4,678 ESA-treated patients were identified (IMPACT—EPO: 991, DARB: 689; Medicare—EPO: 1,788, DARB 1,210). Age and gender distributions were similar between the 2 groups (Mean age: IMPACT: 63.9 vs. 63.2 yrs; Medicare: 74.7 vs. 74.2 yrs, P=NS; % women: IMPACT: 49% vs. 54%; Medicare 59% vs. 62%, P=NS). ESA treatment duration was also similar for EPO and DARB patients. The mean cumulative dose [SD] was 164,786 [175,453] and 244,718 [304,839] Units for EPO and 694 [690] and 989 [1,118] mcg for DARB, resulting in dose ratios of 237:1 and 247:1 for IMPACT and Medicare data, respectively. Based on the recent utilization of ESAs, cumulative cost was 44% and 38% higher for DARB than EPO (IMPACT—EPO $2380; DARB $3427; Medicare—EPO $3534; DARB $4888). After adjusting for covariates, cumulative drug costs remained significantly higher for DARB.  CONCLUSIONS: Based on large health insurance claims databases, this observational study of recent ESA utilization in CKD patients not on dialysis reported dose ratios of 237:1 and 247:1 and cost premiums of 44% and 38% associated with DARB.