SummaryWe systematically reviewed the observational associations of anaemia with outcomes and the effects of interventions to increase haemoglobin concentrations following hip fracture in older people. Anaemia on hospital admission was associated with increased mortality, relative risk 1.64 (95% CI 1.47-1.82), p < 0.0001. After adjustment for co-morbidities, the association of anaemia with increased mortality remained in four of eight observational studies. There was no association of postoperative transfusion with mortality after adjusting for covariates. Transfusion at 80 g.l À1 vs 100 g.l À1 increased acute myocardial infarction, relative risk 1.67 (95% CI 1.01-2.77), p = 0.05. Transfusion threshold was not associated with differences in other outcomes. There were insufficient high-quality studies to inform pre-operative blood transfusion or the use of peri-operative iron or erythropoietin. Studies for most interventions recruited too few participants to determine effects on infections, mortality or function.