Intravenous (IV) iron can decrease transfusion requirements in selected patients with low, normal and moderately elevated ferritin. Whether the syndrome of iron-restricted erythropoiesis (IRE), diagnosed by iron studies, identifies critically ill patients at risk for subsequent red blood cell (RBC) transfusion, and hence, provides a simple method to determine response to IV iron therapy, is uncertain. We aimed to describe the characteristics of patients with IRE on admission to intensive care and determine the optimal variables to identify patients at risk of RBC transfusion who may benefit from early administration of IV iron. The study included 201 consecutive ICU admissions from a single 23-bed combined medical/surgical ICU. The prevalence of IRE on admission to ICU, defined according to ferritin <300 μg/l and transferrin saturation <20%, was 26.2% (95% CI 19.9 to 32.4). The proportion of patients with IRE subsequently receiving RBC transfusion was significantly lower than the proportion of patients without IRE receiving RBC transfusion (absolute mean difference 18.9% [95% CI 4.7 to 33.1, P <0.001]). IRE was not independently associated with risk of transfusion on multivariate analysis, however, a prognostic model with three risk factors (RBC transfusion prior to ICU admission, Hb <100 g/l and ICU length of stay >3 days), had good discrimination and calibration for predicting transfusion (receiver operator curve area under the curve 0.87 [95% CI 0.79 to 0.94, P=0.88], Hosmer–Lemeshow 6.21; P=0.1). Excluding iron overload and using simple prognostic criteria to identify patients at high risk of RBC transfusion may be a preferable strategy for identifying critically ill patients who may benefit from IV iron.