BACKGROUND: Severe anemia in children is a major public health problem in sub-Saharan Africa. In this study we describe clinical and operational aspects of blood transfusion in children admitted to Coast Provincial General Hospital, Kenya. STUDY DESIGN AND METHODS: This was an observational study where over a 2-year period, demographic and laboratory data were collected on all children for whom the hospital blood bank received a transfusion request. Clinical data were obtained by retrospective review of case notes over the first year. RESULTS: There were 2789 requests for blood for children (median age, 1.8 years; interquartile range [IQR], 0.6-6.6 years); 70% (1950) of the samples were crossmatched with 85% (1663/1950) issued. Ninety percent (1505/1663) were presumed transfused. Median time from laboratory receipt of request to collection of blood was 3.6 hours (IQR, 1.4-12.8 hr). Case notes of 590 children were reviewed and median pretransfusion hemoglobin level was 6.0 g/dL (IQR, 4.2-9.1 g/dL). Ninety-four percent (186) were transfused “appropriately” while 52% (120) were transfused “inappropriately.” There was significant disagreement between the clinical and laboratory diagnosis of severe anemia (exact McNemar's test; p < 0.0001). Antimalarials were prescribed for 65% (259) of children who received blood transfusions but only 41% (106) of these had a positive blood film. CONCLUSION: In this setting, clinicians often order blood based on the clinical impression of “severe anemia.” This has implications for laboratory workload and the blood supply itself. However, the majority of children with severe anemia were appropriately transfused. The use of antimalarials with blood transfusions irrespective of blood film results is common practice.