We found that the hand-grip strength, gait speed, premorbid ADL score and ADL score at discharge were not significantly different between anaemic and non-anaemic hospitalized geriatric patients. After adjustment for age, sex, BMI, MMSE, eGFR, number of comorbidities and clinical diagnosis, the means for the ADL score, hand-grip strength and gait speed were similar between patients divided into those with severe, moderate and without anaemia. In addition, our data add some evidence to the importance of severe anaemia as a significant risk factor for the 1 year mortality after discharge. Anaemia is a common condition in hospitalized geriatric patients and its prevalence varies widely depending on the population studied and the different diagnostic criteria [
1‐
6]. In this study, 48 % had anaemia according to the WHO criteria and this is in accordance with other studies [
2,
4,
18]. Anaemia has been demonstrated to be significantly associated with a number of negative outcomes such as an increased mortality and hospitalization, poorer physical and cognitive performance, diminished quality of life, increased frailty and number of falls [
7‐
9,
19‐
23]. Most studies were performed in community-dwelling older patients and the specific relationship between anaemia and muscle strength and gait speed have been investigated in a few studies. Penninx et al. and Haslam et al. found that anaemia was associated with a lower hand-grip strength and leg strength [
7,
11]. In another study, self-reported mobility difficulty prevalence was lowest at haemoglobin concentrations around 14 g/dl and increased as the haemoglobin levels decreased toward 8 g/dl in older community-dwelling women [
24]. An association between anaemia and functional disability, mostly basic and instrumental activities of daily living, has been shown in some studies [
10,
25‐
28] but was less clear or absent in other studies [
11,
23,
29,
30] and Maraldi et al. demonstrated that haemoglobin levels were positively associated with the likelihood of recovery from ADL disability in hospitalized older persons [
4]. Most of the data are epidemiologic in nature and do not prove a causal relationship between anaemia and the different poor health outcomes. Although the underlying mechanisms that might explain the possible association between anaemia and physical decline are unclear, some hypotheses dealing with a diminished oxygen delivery to specific organs that jeopardize their functional capacity (i.e. the cardiorespiratory, musculoskeletal and neurological system) or anaemia as an of important marker of inflammation, have been put forward [
31,
32]. However, our data do not support an association between anaemia and physical performance (hand grip strength and gait speed) and functionality (ADL). A possible explanation might be the patient selection in different studies. The mean age of our study population is higher than in most other studies and most patients admitted to the acute geriatric ward are by its nature frail or prefrail which increases the functional dependency as demonstrated by the fact that only a very small minority had a gait speed higher than 0.8 m/s. Some specific and commonly accepted anaemia-related symptoms and diseases such as fatigue and weakness but also chronic kidney disease, cancer, stroke, infectious and inflammatory disorders are also common in non-anaemic hospitalized older patients and could contribute to some extent to a diminished functionality and a more sedentary lifestyle. Although anaemic patients had more comorbidities in our study, the mean CRP as a marker of inflammation was comparable between the anaemic and non-anaemic group and the prevalence of falls and fractures was even higher in the latter group. All this might explain, at least partially, why we could not demonstrate that anaemia upon admission was a significant risk factor for functional disability, lower gait speed and muscle strength in hospitalized older patients because the majority of the patients were too sick, whether they were anaemic or not. Another explanatory factor could be that the functional assessments vary in a more dynamic way in hospitalized patients due to acute illnesses and their recovery as compared to non-hospitalized subjects. The mean gait speed and handgrip strength in our population were much lower as compared to non-hospitalized older patients but our results correspond well with other studies investigating hospitalized older persons [
33,
34]. Unfortunately, data about the functional recovery after the index hospitalization in our patients are lacking. Our data also demonstrate that severe anaemia is a strong independent risk factor associated with an increased 1 year mortality after discharge and this is in accordance with the results of prior studies in community-dwelling older persons [
9,
19,
20].
We are aware that our study has several limitations. We investigated a selected group of hospitalized older patients and chose some major and frequently used confounding variables, excluding other potential explanations linking anaemia with functionality. As a consequence, our results cannot be generalized to other patient groups. The extensive assessment of functionality in geriatric hospitalized patients is a time-consuming and complex procedure. Moreover, the handgrip strength and gait speed are both focused on the musculoskeletal system and require a minimum of the patients’ cooperation and physical skills which are often lacking in this vulnerable population. This limits the participation rate and as a consequence, the appropriate interpretation of gait speed was available in only 44 %. We are aware that the number of participants is limited but it is inherent to this population that the cooperation is also limited and the dropout high. Finally, it was not our intention in this study to diagnose the specific underlying cause for the anaemia (acute or chronic, anaemia of inflammation, iron deficiency anaemia etc.) but it is possible that the results of the geriatric assessment and the prognostic significance of anaemia differ according to the underlying anaemia aetiology.