Practice guidelines for use of an erythropoietic agent (EpA) have been previously developed for patients with all cancers, but these have not specifically addressed non-myeloid hematological malignancies. Given issues of benefit, cost, access to treatment and practice variation for these patients, the Cancer Care Ontario Program in Evidence - Based Care (CCO PEBC) conducted a systematic review and has developed a practice guideline. Entries to MEDLINE, CANCERLIT, EMBASE, the Cochrane Library databases, and abstracts of the American Societies of Clinical Oncology (ASCO) and Hematology (ASH) were searched. A hierarchy of outcomes was developed that included survival, quality of life (QoL), transfusion requirements and improvements in hemoglobin concentration/hematocrit. To validate conclusions and recommendations, the practice guideline was sent to Ontario practitioners in July 2006; responses are now being collated. Eighteen reports (14 articles, 4 abstracts) of randomized trials met eligibility criteria. None of the 3 trials reporting survival outcomes detected a benefit with use of an EpA. Of 7 trials evaluating QoL, 6 reported superior outcomes in patients receiving an EpA. None of those trials sufficiently reported methodologic parameters required by proposed guidelines for analyzing, interpreting and reporting QoL measures. Use of an EpA significantly reduced the proportion of patients transfused in 5 trials evaluating this outcome; reductions ranged from 15% to 40%. The absolute risk reduction ranged from 15% to 24%; the number needed to treat to prevent a transfusion ranged from 4 to 6. Use of an EpA was not associated with a reduction in the mean/median number of units transfused. In 10 studies, either a statistically significant increase in the hemoglobin concentration/hematocrit or in the proportion of patients with an increase in the hemoglobin concentration/hematocrit was seen. For this practice guideline, we interpreted the evidence as providing insufficient data to justify use of an EpA in order to improve survival or QoL. However, there are compelling data showing that use of an EpA reduces the risk of requiring a transfusion. Initiatives such as the Commission of Inquiry on the Blood System in Canada (“the Krever Commission”), state that alternatives to transfusion be offered to patients because of associated known and potentially unknown adverse consequences of blood products. Thus, the use of an EpA is recommended as an alternative to transfusion. However, the decision to use an EpA to reduce transfusion requirements should primarily consider individual patient values and the likelihood that a patient will require a transfusion so that patients are not exposed to unnecessary treatment and the health care system to additional costs.