Limited data exist on the epidemiology, evaluation, and prognosis of otherwise unexplained anaemia of the elderly in heart failure (HF). Thus, we aimed to determine the incidence of anaemia, to characterize diagnostic testing patterns for potentially reversible causes of anaemia, and to evaluate the independent association between incident anaemia and long-term morbidity and mortality. Within the Cardiovascular Research Network (CVRN), we identified adults age ≥65 years with diagnosed HF between 2005 and 2012 and no anaemia at entry. Incident anaemia was defined using World Health Organization (WHO) haemoglobin thresholds (<13.0 g/dL in men; <12.0 g/dL in women). All-cause death and hospitalizations for HF and any cause were identified from electronic health records. Among 38 826 older HF patients, 22 163 (57.1%) developed incident anaemia over a median (interquartile range) follow-up of 2.9 (1.2-5.6) years. The crude rate [95% confidence interval (CI)] per 100 person-years of incident anaemia was 26.4 (95% CI 26.0-26.7) and was higher for preserved ejection fraction (EF) [29.2 (95% CI 28.6-29.8)] compared with borderline EF [26.5 (95% CI 25.4-27.7)] or reduced EF [26.6 (95% CI 25.8-27.4)]. Iron indices, vitamin B12 level, and thyroid testing were performed in 20.9%, 14.9%, and 40.2% of patients, respectively. Reduced iron stores, vitamin B12 deficiency, and/or hypothyroidism were present in 29.7%, 3.2%, and 18.6% of tested patients, respectively. In multivariable analyses, incident anaemia was associated with excess mortality [hazard ratio (HR) 2.14, 95% CI 2.07-2.22] as well as hospitalization for HF (HR 1.80, 95% CI 1.72-1.88) and any cause (HR 1.77, 95% CI 1.72-1.83). Among older adults with HF, incident anaemia is common and independently associated with substantially increased risks of morbidity and mortality. Additional research is necessary to clarify the value of routine evaluation and treatment of potentially reversible causes of anaemia.