1. Introduction
According to Statistics Austria, in 2005, there were 1,307,945 people in Austria over the age of 65 years, representing a percentage of around 15%. By 2020, this number had risen to 1,707,643 and already accounted for 19.1% of the total population. By 2030, according to demographic forecasts, 2,140,717 and by 2040 already 2,490,916 persons over 65 years of age—that is 23.3% and 26.4% of the population, respectively—will be living in Austria [
1]. In addition, the number of inpatients in nursing homes in Austria aged 65 and over increased by a full 25% between 2015 and 2020, from 75,632 to 95,263 [
1].
Primarily elderly people in institutions are among the high-risk groups for the development of malnutrition due to various age-related changes. Aging-related conditions such as dementia [
2], sensory loss, loss of appetite [
3], inflammation [
4,
5,
6], comorbid illnesses, dysphagia, and polypharmacy increase the likelihood of weight loss or poor nutrition [
7,
8,
9,
10]. Recently published studies have confirmed that manifest malnutrition, regardless of definition, leads to increased mortality (2% in hospitals and 6% in nursing homes) [
11] and, with a median of 6 days, a significantly prolonged length of stay (LOS) as well as an increase in readmission rate. Thus, it causes more costs than compared with well-nourished patients with a similar diagnosis [
11,
12,
13,
14]. For example, according to a study published in 2003, malnourished patients stayed in the hospital for 16.7 days (SD: 24.5), whereas normally nourished patients had a length of stay of 10.1 days (SD: 11.7). Hospital costs for malnourished patients were increased by up to 308.9% [
15]. Schmidt et al. estimate that in Germany, costs of about 9 billion euros per year are caused by nursing measures or hospitalizations due to malnutrition [
16]. S. Moramarco et al., in a study published in 2020, reported a prevalence of 67.4% of hypoalbuminemia in hospitalized patients aged 65 years or older, highlighting the role of albumin as a useful marker of malnutrition and frailty [
17].
In 2019, X. Hong et al. also confirmed that elderly hospitalized patients who were well-nourished and had higher serum levels of total protein and albumin were less likely to become frail [
18]. Furthermore, in a large-scale frailty study by Song et al., it was shown that hypoalbuminemia is a prognostic mortality factor in the elderly, whether they live in the community, are hospitalized or institutionalized. Accordingly, low albumin levels are associated with poorer recovery after acute pathologies and additionally increase the risk of death by 57% in individuals aged 65–102 years [
19]. In view of the demographic development of the elderly population in the coming years and decades, the high relevance of the problem becomes apparent.
Especially in the central European region, the role of malnutrition in nursing homes has only been studied to a limited extent and is mostly significantly underestimated [
1,
20].
The aim of this study was to demonstrate the prevalence of malnutrition in the patient population of an Austrian nursing home and shed light on its clinical relevance. The hypothesis was that undernourished patients have an increased susceptibility to the occurrence of complications as well as an increased mortality rate.
2. Materials and Methods
The chosen methodological approach was a non-experimental, descriptive, quantitative cross-sectional design. The data for this study arises from the database of a large nursing home in Vienna, Austria, which has not been evaluated or analyzed so far.
Data from patients admitted to the nursing home between 01/2017 and 08/2020 were used for this analysis.
Exclusion criteria in the study were an inpatient stay of less than 12 months, no admission blood samples, underlying malignancies, infectious disease, and surgery performed during the inpatient stay, as well as patients younger than 65 years of age.
The patient data were submitted in already anonymized form and included only data of “older/elderly” patients with at least 65 years of age. Using the case numbers already anonymized on-site, the patient data was sorted chronologically. From the extensive database of 545 patients, 120 patients were eligible for this study. Age, sex, height, and weight at the time of admission were entered into the resulting register. In addition, if available, weight was also entered 3, 6, and 12 months after the time of admission to calculate the BMI and to determine any changes in weight and malnutrition prevalence. Laboratory parameters of the admission blood were used to further determine the health and nutritional status of each nursing home resident. This analysis included the testing of C-reactive protein (CRP), transferrin, serum albumin, and hemoglobin. Finally, the status of the nursing home residents was noted. The status of the individuals indicated whether they were deceased, discharged, or still in residential care at the time of data collection. The malnutrition status was calculated using the guidelines of the German Society of Nutritional Medicine (DGEM). The prevalence was noted after 3, 6, and 12 months of stationary stay.
Statistical Analysis
The open-source statistical program package R (version 4.0.4 x86) and the statistical program IBM SPSS Statistics 27 under Windows 10 were used for the analysis.
The alpha error level was set at 5% (two-sided) for each test; no adjustment was made for the alpha error. The results of the statistical tests are therefore to be understood as purely descriptive. Missing values were not replaced. The open-source statistical software package R (version 4.0.4 x86) under Windows 10 was used for the evaluation.
For unrelated categorical variables such as sex, Fisher’s exact test (Fisher) (2 × 2 tables) was used.
For unrelated ordinal variables in the case of two independent groups, such as age, BMI, and laboratory parameters, the Mann–Whitney–exact U (MWU) test was used.
Additionally, for the comparison of two independent groups, the two-sample
t-test for independent samples was used. In the case of variance heterogeneity (checked by Levene’s test), Welch’s
t-test was applied. In the case of a non-normal distribution of the data (test by Kolmogorov–Smirnov test with Lilliefors correction, alpha = 10%), the exact Mann–Whitney–U test was carried out (
Table 1).
5. Discussion
In this study, a very high mortality rate of 45.8% was observed within a period of 3 years and 7 months. In comparable studies, the median survival time was 2.2 years, while the 3-year mortality rate in comparable studies was around 50% [
35].
An exceptionally high rate of malnourished nursing home residents was found, with a percentage of 37.9% within the first 3 months of admission and 41.1% after 6 months, respectively. However, according to the DGEM criteria, only 32.6% were considered to be malnourished after one year.
When it comes to malnutrition, differences in the patient populations as well as the lack of uniform definitions and the absence of generally valid diagnostic criteria hinder detailed comparability to other international studies. However, most comparable studies indicate prevalence rates of around 15–45% [
36,
37,
38], while other experts mention prevalence rates of malnutrition in nursing homes of up to 60% [
39]. In this study, a definition of the DGEM guidelines was used, which is based on purely anthropometric data, such as BMI and weight history [
40]. However, although weight and its progression are a very important indication of malnutrition that can be examined quickly, it is a much more complex subject.
Uniform diagnostic criteria for malnutrition, which additionally include laboratory data, are therefore indispensable.
Presumably, due to the small number of people included, no correlation with mortality could be shown, as was often the case in comparable studies [
41,
42]. More than half of nursing home residents were found to have low serum albumin saturation. Most comparable international studies performed in nursing homes describe prevalence rates of hypoalbuminemia between 19% and 55% [
21,
26,
27,
37]. Especially in recent work, the role of serum biomarkers, particularly serum albumin, in diagnosing or monitoring malnutrition is controversial, mainly due to its lack of specificity and long half-life (approximately 20 days) [
43,
44]. Thus, a connection between malnutrition and hypoalbuminemia was not found in this study.
Hypoalbuminemia may be the result of decreased production of albumin or increased loss of albumin via the kidneys, gastrointestinal tract, skin, or extravascular space. It may also be the result of increased catabolism of albumin, for instance, in inflammation, or a combination of two or more of these mechanisms [
45]. In addition to the development of the disease itself, the condition of hypoalbuminemia could also be a consequence of the aging process, as albumin levels decrease with age [
26].
However, the high prevalence of low albumin in elderly nursing home residents is particularly worrying, as this study also showed a clear correlation between hypoalbuminemia and increased mortality. Serum albumin remains the most cost-effective, common, and practical protein for assessing protein status and response to nutritional support and should therefore never be disregarded as part of a nutritional assessment [
25].
Transferrin has also often been used as a marker of nutritional status in the recent past [
46]. Serum levels decrease in the setting of severe malnutrition, but this marker is unreliable in the assessment of mild malnutrition in several recent studies [
44]. According to the results of this work, transferrin does not serve as a suitable marker for nutritional status. However, transferrin is an important correlate of mortality in elderly patients [
23], which can also be supported by the results of this study. According to the results of this study, hypoalbuminemia and low transferrin levels should be considered dangerous and problematic conditions on their own, especially in nursing home residents.
As described above, the etiology of malnutrition as well as the clinical presentation are diverse. Inflammation and emaciation are often observed simultaneously in the elderly, which is called malnutrition-inflammation-cachexia syndrome [
47]. However, no significant association between malnutrition and elevated CRP values was found in this study.
Equally noticeable were the high prevalence rates of anemic nursing home residents (48.8% of females, and 70.6% of male residents). Izaks et al. showed that anemic individuals aged 86 and over had a higher 5-year mortality rate than those with normal hemoglobin levels [
48].
The most common incidence of anemia in older adults is related to malnutrition and is due to iron, folate, and/or vitamin B12 deficiency [
49].
Of course, the laboratory parameters may be influenced by underlying chronic diseases; however, Zhiying Zhang et al. could show that BMI and several blood biochemicals, including albumin and hemoglobin, are useful biomarkers for adult malnutrition, even in the presence of chronic inflammation [
50]. However, when analyzing malnutrition, laboratory parameters should always be considered in relation to anthropometric data and unplanned weight loss over time [
51].
Finally, it is to be said that the difficulty of finding a generally valid definition is not only because malnutrition rarely is the primary reason for admission but also due to the high individual heterogeneity of malnutrition, nutrient deficiencies, or metabolic and nutritional disorders, especially in the elderly.
Therefore, physicians play an important role, not only in the early detection and diagnosis of malnutrition, monitoring and possible therapy but also in the clinical application of personalized nutritional support for malnourished residents in the nursing home. Patients at particular risk of malnutrition with functional and/or cognitive impairment or dementia, swallowing problems, and depression can be identified and cared for separately by specially trained staff [
41,
52].
Limitations
The findings of this study have to be seen in light of some limitations.
Firstly, the sample size is relatively small, with 120 randomly selected patients who have been inpatients in the nursing home for at least one year. A larger sample size could definitely increase the power of similar studies.
A further major weakness of this study is the absence of laboratory data or inadequate collection or recording of anthropometric data. This is because patients are not routinely admitted by a doctor, as is usual in a hospital.
In addition, the data on the cause of death could not be collected in this study, therefore, a more precise evaluation and contrast were not possible.
In most comparable studies today, the GLIM criteria are used to assess malnutrition in nursing homes [
53]. In the present paper, these could not be carried out due to a lack of body composition measuring techniques.