Servicios personalizados
Inglés (pdf)
Articulo en XML
Referencias del artículo
Traducción automática
Enviar articulo por email
Citado por SciELO 
Citado por Google 
Similares en
SciELO 
Similares en Google 
Permalink
INTRODUCTION
The identification of nutritional risk in elder patients is becoming more and more important, considering that a high nutritional risk may independently be associated with a longer hospital stay (1).
Since 1999, Ferguson et al. (2) developed a Malnutrition Screening Tool – MST – in acute hospitalized adult patients in order to identify patients at high nutritional risk and requiring nutritional therapy. The assessment of such tool included 408 patients admitted to an Australian hospital, excluding pediatric, maternity, and psychiatric patients. The study's target population was asked two questions related to appetite and recent, involuntary weight loss, showing a reliability between 93 % and 97 % for the high malnutrition-screening tool amongst evaluators.
After the application of nutritional screening tools in hospitalized populations of older adults, a prevalence between 15 % and 60 % (3-5) has been reported in the literature. In a study carried out by Stratton et al. (6), 150 patients of advanced age who were consecutively admitted to a health institution were prospectively evaluated. These patients underwent nutritional screening using the MUST tool, and 58 % of the study's target population was found to be at risk for malnutrition; these individuals showed a greater rate of in-hospital mortality, as well as longer hospital stays when compared to those other patients who were identified as at low risk. A similar finding was reported by Matins et al. (1), whose investigation in older adult patients classified as malnourished using the NRS-2002 screening tool found that their subjects had a greater risk of a longer stay (> 8 days).
Recently, nutritional screening tools have been articulated and added to electronic medical records, showing evidence that implementation of an electronic-format tool for detecting malnutrition improves the knowledge, approachability, and practice of the healthcare staff in hospitals. This creates interactions in multidisciplinary nutritional care and, therefore, results in timely referral to specialized nutritional management (7).
As a result, the use of a nutritional risk detection tool has a short-term, positive effect on the quality of a hospital's nutritional care process, since detection of the risk of malnutrition represents a crucial starting point for the successful management of malnourished patients. Eglseer et al. (8) suggest the importance of joining efforts in order to sustainably maintain the performance of nutritional screening, reinforcing a positive standpoint, designating motivated "key opinion leaders", and ensuring the consented support from the management within the framework of comprehensive healthcare.
Given the importance of the topic and the implications for clinical practice and the health of older adult patients, the objective of our research was to establish the prevalence of the risk of malnutrition in the patient population of a high-complexity level hospital, together with its associated factors.
MATERIALS AND METHODS
STUDY DESIGN AND PARTICIPANTS
This was a cross-sectional study where the prevalence of nutritional risk and associated factors was determined. Patients were included of both sexes, aged ≥ 60 years, hospitalized during the first 24 hours of being admitted to the emergency observation area of a high-complexity level healthcare services provider institution in Colombia from January 1 to December 31, 2018. Patients without a nutritional risk measurement were excluded.
DATA COLLECTION
The nutritional risk calculation tool applied in this investigation was developed by Fergusson et al., and is known by the acronym MST (Table I). For this tool, the authors reported a high reliability (between 93 % and 97 %), this also being a fast, simple-to-apply tool with an application time at the institution averaging five minutes per patient (2).
Table I. Malnutrition Screening Tool
*Score ≥ 2 = patients at risk of malnutrition.Source: Ferguson M, Capra S, Bauer J, Banks M. Development of a valid and reliable malnutrition screening tool for adult acute hospital patients. Nutrition 1999;15(6):458-64.
Nutritional screening was carried out in patients who complied with the inclusion criteria. Two registered dietitians/nutritionists carried out a structured interview and a nutritional risk evaluation.
The information was plotted in an Excel (version 2013) database. Data filtering was carried out using simple frequencies and crossing variables. Information was processed by the SPSS statistical package, version 25.0.
STATISTICAL ANALYSIS
The descriptive analysis of qualitative variables used absolute and relative frequencies expressed as percentages. For the quantitative variables, measures of central tendency (mean and median) and dispersion (range and standard deviation) were applied.
The prevalence of nutritional risk was measured as a probability expressed as percentage. The association between risk of malnutrition and different factors (sex, age group, pooled stay, and diagnosis) was assessed using Pearson's chi-squared independence test. Odds ratios (OR) and their respective 95 % confidence intervals were also used. For numerical variables (age and hospital stay), the Kolmogorov-Smirnov and Shapiro-Wilk tests were used for the previous testing of normality. In case of normality, Student's t-test was used for the mean differences between two independent groups, with a prior evaluation of the homogeneity of variances (Levene's test), and for distributions other than normal a non-parametrical Mann Whitney-Wilcoxon (M-W) test was used.
A multivariate analysis was performed by means of unconditional logistic regression for risk of malnutrition, estimating the ORs and 95 % confidence intervals. The statistical testing was evaluated at a 5 % significance level (p < 0.05).
ETHICAL DISCLAIMER
This study was approved by the Méderi Technical Research Committee (CIMED, for its acronym in Spanish) and the Universidad del Rosario Research Ethics Committee, ensuring respect for patient confidentiality guidelines, and complying with the Declaration of Helsinki and the national regulations intended to guarantee the ethical principles for medical research in humans.
RESULTS
The agreement measurement for the malnutrition risk scale was evaluated by two nutritionists, who found a high level of concordance (kappa = 0.85, p < 0.001).
A total of 7192 patients, predominately female, with a minimum age of 61 years and a maximum age of 108 years, comprised the total cohort. Average age was 77.1 ± 9.2 years, and median age was 77 years. The predominant age group was that of 70-79 years (Table II). The five most frequent pathologies for hospital admission were urinary tract infections, congestive heart failure, chronic obstructive pulmonary disease with an acute exacerbation, bacterial pneumonia, and gastrointestinal hemorrhage (Table II). The pathology groups, according to the 10th revision of the International Classification of Diseases and Related Health Problems (ICD-10), reported with the highest frequency were circulatory system diseases (1594; 22.2 %), respiratory system diseases (1105; 15.4 %), and genitourinary tract diseases (794; 11 %).
The prevalence of the risk of malnutrition was 41.4 % (n = 2955), and was significantly higher in women (42.5 % vs 39.9 % in men, p = 0.0028). A significant linear tendency was evidenced in the study population: risk of malnutrition increased with older age (p < 0.001, tendency by chi-square test) and with extended hospital stay (p < 0.001, tendency by chi-square test) (Table III). The average hospital stay was 7.78 ± 7.36 days (median = 5.79), and was significantly longer in patients at higher nutritional risk (8.64 ± 8.05 days, median = 6.47) than in those classified at low nutritional risk (7.17 ± 6.76 days; median = 6.47; p < 0.001, M-W).
Table III. Relationship between age and length of stay (pooled) with risk of malnutrition, measured with MST
A significant association between diagnosis (ICD-10) and risk of malnutrition was found (p < 0.001, Pearson's chi-square test), with the highest prevalence for the risk of malnutrition being found in the diagnoses of non-specified delirium (56.4 %), specified pneumonia (50.8 %), diarrhea and gastroenteritis of suspected infectious etiology (47.8 %), non-specified gastrointestinal hemorrhage (47.6 %), high blood pressure (44.4 %), non-specified urinary tract infection (43.6 %), non-specified bacterial pneumonia (42.7 %), and other abdominal pain (41.9 %).
MULTIVARIATE ANALYSIS
The factors that together were significantly linked to high risk of malnutrition were increased hospital stay, and increased age (ICD-10). The diagnoses with a stronger association with risk of malnutrition were: non-specified delirium (OR = 5.98, 95 % CI: 2.78 to 12.86), diarrhea and gastroenteritis of suspected infectious etiology (OR = 5.01, 95 % CI: 2.44 to 10.32), non-specified gastrointestinal hemorrhage (OR = 4.44, 95 % CI: 2.38 to 8.28), specified pneumonia (OR = 4.43, 95 % CI: 2.11 to 9.30), high blood pressure (3.94, 95 % CI: 2.07 to 7.50), other abdominal pain (OR = 3.80, 95 % CI: 1.81 to 7.99), non-specified urinary tract infections (OR = 3.64, 95 % CI: 2.07 to 6.24), non-specified acute bronchitis (OR = 3.22, 95 % CI: 1.56 to 6.65), and non-specified bacterial pneumonia (OR = 3.02, 95 % CI: 1.65 to 5.55) (Table IV).
DISCUSSION
Out of the 7192 patients receiving nutritional screening, 2955 cases at risk of malnutrition were detected, i.e., a prevalence of 41.4 % was found, similar to that reported in previous studies carried out in Venezuela (48.4 %) (9), Mexico (40.8 %) (10), Ecuador (37.1 %) (11), and Colombia (60.1 %) (12). The prevalence of malnutrition in studies using the MST tool was equivalent to that reported in this study. For example, a university hospital in southern Brazil reported malnutrition in 33.1 % of the population (13), while in two primary care university hospitals in Porto the estimated prevalence was 55.1 % (1).
According to a systematic review by Correia et al. (14), which included 66 studies involving 29,474 patients from 12 Latin American countries, a prevalence of malnutrition was noted in 40 %-60 % of patients at the time of admission; a prevalence greater than 45 % was specifically reported for Colombia (15,16).
Regarding hospital stay, it is considered that the longer patient stay is, the higher the risk of malnutrition becomes. Martins et al. (1) reported that patients classified as malnourished were at greater risk, regardless of longer stay (8 days). These values were also reported by prevalence studies suggesting that the prevalence of malnutrition increases with stay duration (11,17,18).
The average hospital stay in this study was 7.78 days, with a stay significantly higher in patients at high nutritional risk lasting 8.64 days. Peniche-Herrera et al. (19) evaluated 138 medical files seeking to determine whether the risk of preoperative malnutrition was a causative factor for prolonged hospital stay after gastrointestinal surgery, with evidence that the presence of the risk of preoperative malnutrition is linked to prolonged hospital stay (OR = 1.33, 95 % CI: 1.07-1.64, p = 0.008).
A population study in patients admitted to hospital in two European countries reported that in patients at high risk of malnutrition stay was longer than 11 days (20). A similar report was made by Sorensen et al. (21) in the EuroOOPS study, which affirmed that patients at risk of malnutrition had more complications, greater mortality, and longer stays than patients at no risk. Therefore, hospital stay is a key factor in the administrative management of healthcare institutions, considering that malnutrition is common at hospital admittance and tends to worsen during hospitalization.
Out of the ten pathologies with the largest number of patients based on the ICD-10 diagnosis, there was evidence that circulatory and respiratory system diseases had a prevalence of risk of malnutrition higher than 30 %. Patients with congestive heart failure had a risk of malnutrition of 33.9 %, a value close to that reported by Gomes et al. (22), with high nutritional risk in 44 % of patients with heart failure. A prevalence of 36.5 % and 23.2 % for the risk of malnutrition was reported for pathologies such as chronic obstructive pulmonary disease (COPD) with or without exacerbation, respectively. These figures are similar to those reported in the literature, ranging between 10 % and 60 % (23,24), with the development of nutritional intervention strategies being of paramount importance in clinical practice because malnutrition is common amongst COPD patients (25).
Other relevant pathology in this study was the diagnosis of high blood pressure, where 32.5 % were at risk of malnutrition. This value is lower than that reported in the research carried out in Colombia by Giraldo et al. (26), who found that 62.5 % of older adult patients with high blood pressure presented risk of malnutrition. The differences in cohort conformation, the different screening tools used to establish the risk of malnutrition, and the particularities of each hospital institution may explain the differences found in the studies discussed in this publication.
The present study has various strengths. The size of the sample stands out with a large volume for the study population, an adequate representation of both sexes and age groups, and the inclusion of different health insurers. In addition, information was recorded by nutritionists, thus creating acceptable confidence at the collection of the data defined by the MST tool, established in the electronic medical records.
Finally, with regard to the implications for research and clinical practice, the present research suggests there is a need to carry out additional studies evaluating nutritional risk at different hospitals in the country, given the population's variability. Likewise, the importance of identifying and characterizing the clinical outcomes linked to the risk and cost of malnutrition is highlighted. This is in pursuit of the development of nutritional care programs by hospital institutions, taking into account the factors that increase the risk and supported by the available evidence.
One in every two patients admitted to a hospital institution had a high risk of malnutrition linked to hospital stays longer than 8 days, age ≥ 80 years, and the specific diagnoses of delirium, diarrhea and gastroenteritis of suspected infectious etiology, gastrointestinal hemorrhage, specified pneumonia, high blood pressure, other abdominal pain, non-specified urinary tract infections, acute bronchitis, and bacterial pneumonia.
