Prevalence and risk factors of anaemia in older hospitalised patients

Abstract

Introduction: We aim to investigate the prevalence and risk factors for anaemia among the elderly in a tertiary hospital. Methods: A retrospective study was conducted on 742 consecutive geriatric patients who were admitted to the medical unit of our hospital. The demographic and laboratory values for patients were examined. Anaemiarelated data were studied. Patients with no/missing demographical or laboratory data and those who passed away after admission were excluded. Results: There was a high prevalence of anaemia in the study population. Although there were a significant proportion of nutritional causes, non-nutritional causes formed the majority. Significant risk adjusted risk factors associated with anaemia are male gender, Charlson's co-morbidity index and abnormal serum creatinine levels ≤60ml/min. (OR=1.469 [1.177-1.833], p=0.001; OR=1.147 [1.107-1.189], p<0.001; OR=1.335 [1.064-1.675], p=0.013 respectively.) Contributory factors for sub-types of anaemia were further investigated. A non-Chinese ethnicity, NSAID and anti-platelet use were significant risk factors for iron deficiency anaemia. (OR=2.035 [1.207-3.432], p=0.008; OR=8.202 [3.628-18.540], p<0.001; OR=2.080 [1.260-3.435], p=0.004 respectively.) For B12 deficiency, significant risk factors were non-Chinese ethnicity and metformin use. (OR=1.869 [1.036-3.370], p=0.038; OR=3.314 [1.690-6.499], p<0.001 respectively) A high Charleson's co-morbidity score and male gender correlated with anaemia of chronic disease. (OR=1.192 [1.104-1.288], p<0.001; OR=2.030 [1.198-3.440], p=0.009 respectively.) For anaemia of chronic renal disease, no significant predictors were found. Conclusion: Anaemia remains prevalent in the hospitalised geriatric population. Specific risk factors may help in the diagnosis and management of sub-types of anaemia.