Abstract
Background: Anaemia is associated with adverse outcomes in elderly community-dwelling individuals, but this problem is less well characterised in the inpatient setting. Aims: To determine the prevalence of anaemia and its associations in a well-defined cohort of internal medicine inpatients. Methods: A retrospective cohort study of non-elective admissions under internal medicine at Palmerston North Hospital, New Zealand, was conducted for 4 months of 2008 with outcome analysis on 1 March 2010. Results: At admission, 497 of 1491 (33.3%) patients were anaemic by World Health Organization criteria (haemoglobin <130 g/L for males; <120 g/L for females). Anaemia was more prevalent in males (38.1%) than females (28.2%), P < 0.001, in patients aged 65 years or older (41%) than in those under 65 (21.3%), P < 0.001, in New Zealand Europeans (34.3%) than in Maori and people from the Pacific Islands (26.4%), P= 0.04, and in patients admitted primarily because of malignancy, endocrine/metabolic disease, infection, and acute coronary syndrome/congestive heart failure (P < 0.001). Anaemia was independently associated with increased length of hospital stay (7.3 days vs 5.1 days in non-anaemic patients; P < 0.001), with mortality (P < 0.001) and unplanned hospital readmission (P < 0.001) during the follow-up period. Anaemia was infrequently acknowledged or investigated. Secondary analysis using a haemoglobin threshold of 110 g/L showed similar results. Conclusions: Anaemia is highly prevalent among medical inpatients with variation because of gender, age, race and reason for admission. Anaemia independently predicts for prolonged hospital stay, increased mortality and shorter time to readmission, but is usually not documented or investigated in this setting.