Study objective This study investigated in a large sample of in-patients the impact of mild-moderate-severe anaemia on clinical outcomes such as in-hospital mortality, re-admission, and death within three months after discharge. Methods A prospective multicentre observational study, involving older people admitted to 87 internal medicine and geriatric wards, was done in Italy between 2010 and 2012. The main clinical/laboratory data were obtained on admission and discharge. Based on haemoglobin (Hb), subjects were classified in three groups: group 1 with normal Hb, (reference group), group 2 with mildly reduced Hb (10.0–11.9 g/dL in women; 10.0–12.9 g/dL in men) and group 3 with moderately-severely reduced Hb (<10 g/dL in women and men). Results Patients (2678; mean age 79.2 ± 7.4 y) with anaemia (54.7%) were older, with greater functional impairment and more comorbidity. Multivariable analysis showed that mild but not moderate-severe anaemia was associated with a higher risk of hospital re-admission within three months (group 2: OR = 1.62; 95%CI 1.21–2.17). Anaemia failed to predict in-hospital mortality, while a higher risk of dying within three months was associated with the degree of Hb reduction on admission (group 2: OR = 1.82;95%CI 1.25–2.67; group 3: OR = 2.78;95%CI 1.82–4.26) and discharge (group 2: OR = 2.37;95%CI 1.48–3.93; group 3: OR = 3.70;95%CI 2.14–6.52). Normocytic and macrocytic, but not microcytic anaemia, were associated with adverse clinical outcomes. Conclusions Mild anaemia predicted hospital re-admission of older in-patients, while three-month mortality risk increased proportionally with anaemia severity. Type and severity of anaemia affected hospital re-admission and mortality, the worst prognosis being associated with normocytic and macrocytic anaemia.
