Background/Aims: In a cohort of hemodialysis patients, we evaluated the hypothesis that weekly administration of intravenous (IV) darbepoetin-α (DA) was associated with lower total erythropoiesis-stimulating agent (ESA) requirements as compared to a regimen of multiple subcutaneous (SC) doses per week of epoetin-β (EB). Methods: We studied 1,159 hemodialysis patients who were treated exclusively with either IV DA or SC EB across a network of Portuguese clinics during 2004. Linear regression was used to assess the adjusted relationship between the ESA regimen and weekly ESA requirements over the period of observation. Generalized estimating equations were applied in order to model the population average effects of the correlated mean weekly ESA dose for each individual. We also calculated propensity scores for the receipt of DA and assessed the relationship between ESA type and dose requirement within each quintile of the score. Results: The adjusted dose of IV DA, when expressed as a proportion of the dose used in EB-treated patients, did not differ from the dose administered to EB recipients (0.961, 95% CI 0.904, 1.021). A similar relationship was observed within each propensity score quintile. Conclusions: Hemodialysis patients who received IV DA had dose requirements that were similar to their counterparts who were treated with SC EB. A once-weekly dosing regimen and avoidance of SC administration enhance the attractiveness of DA as an alternative to traditional ESAs. The potential for unmeasured confounding, restriction to a population that was treated with a single ESA preparation and application of a 200 IU:1 µg EB:DA dose conversion are important limitations of this study.