Elderly Feeding and Nutrition during Hospitalization: An Integrative Review ()
1. Introduction
Human aging is a desirable event and represents an achievement of humanity. It brings with it social chains that exert impacts on the economy, politics and especially on health. It is a progressive and gradual process that implies motor and sensory losses over time, rendering individuals more vulnerable and susceptible to the emergence of diseases that directly affect their functionality [1] .
During aging, it is common for changes in body composition to occur, such as decreased muscle tissue, loss of strength and greater concentration of adipose tissue associated with malnutrition [2] . Such changes can lead older adults to require hospitalization to take care of their health/disease and extend their lives, even though it represents a high risk to their health by itself, especially when it happens for prolonged periods of time. At this moment, there is a high probability of suffering negative clinical outcomes such as incontinence, malnutrition, depression, worsening of morbidities and comorbidity, cognitive decline, deterioration of functional capacity and death [3] .
In the eating process, food must be chewed, transformed into a bolus and taken to the pharynx, propelled mainly by the tongue. This process is highly complex, as it involves many muscles of the oral cavity, larynx and esophagus, and together more than 30 nerves and muscles for volitional and reflexive activities during eating [4] .
Functional losses require health/illness care, and often represent, for the elderly, a trajectory of continuous demand for primary, secondary and tertiary health care services. Therefore, it is the elderly population segment (60 years of age or over) that tends to resort to hospital admissions more frequently. However, it is not pertinent to focus efforts solely on identifying and evaluating physiological factors that affect the nutrition of hospitalized elderly people. The hospital, as a necessary environment for maintaining health and life, also has intrinsic elements in its physical and operational structure that directly affect the oral feeding process of elderly people [5] .
A meta-analysis involved 583,972 older adults from 24 European countries with the objective of assessing the malnutrition risk in different contexts. The malnutrition risk was 28% in hospitals, 17.5% in home-care and 8.5% in the community [6] . For researchers, malnutrition in hospitalized aged people is influenced by the clinical condition, poor absorption of nutrients, use of drugs or increased metabolic demand due to the pathology [7] .
It is observed that the malnutrition risk is significantly higher in hospitalized aged people, which reinforces the importance of carrying out this review with a focus on this population segment. The broad and systematic identification of nutritional and dietary factors that interfere with the nutritional condition are essential for the health team, as they represent support to ensure care related to the nutritional needs of hospitalized aged people.
The search for evidence of the relationship between hospital admission, physical frailty and difficulties with oral feeding in the elderly is relevant. It is essential for the healthcare team to have studies that provide scientific evidence about this relationship, and thus adopt more assertive directions in the practice of gerontological care in the hospital context.
In view of the above, this integrative literature review aimed at searching the current literature for the elements that exert an influence on the diet and nutrition of hospitalized aged people.
2. Method
This is an integrative literature review and, for this purpose, the methodological stages of establishing a research question, sampling or searching the literature, categorizing the studies, evaluating those included in the review, interpreting the results and presenting the review were undertaken [8] .
Identification of the theme and elaboration of the research question fulfilled the first review stage. The PCC acronym was used for structuring the guiding question and elaborating the search strategy of the studies, where P = Population or Patients (aged people); C = Concept (diet, nutrition); C = Context (hospitalization), according to Chapter 11.2.2 of the JBI Manual for Evidence Synthesis [9] . Thus, the following question was structured: “Which is the current panorama of national and international scientific production on the elements influencing elderly feeding and nutrition during hospitalization?”
The following inclusion criteria were defined in the second review stage: targeting people aged ≥ 60 years old, considering that, in Brazil, individuals over 60 years old are considered elderly according to the Elderly Statute (Federal Law no. 10741/2003); with no limit regarding publication date and being directly related to the topic and developed in the hospital context. The exclusion criteria were as follows: materials appearing as editorials, opinion articles, dissertations, theses and review articles seeking only the inclusion of original articles.
Also, in the second stage of this review, the specific search strategy for each portal and database was elaborated. Initially, the Medical Subject Headings (MeSH) descriptors was used (Aged, Frailty, Frail Elderly, Inpatients, Hospitalization, Elderly Nutrition, Eating) and, subsequently, the translation for the specific descriptors accompanied by Boolean operators (AND, OR), as shown in Table 1.
The search strategy was applied by the main researcher in the Medical Literature Analysis and Retrieval System Online (MEDLINE) (PubMed Portal); Biblioteca Virtual em Saúde (BVS); Embase, Cumulative Index to Nursing and Allied Health Literature (CINAHL) and Scopus databases in May 2023. Table 1 shows the search strategies.
The articles were organized in the third stage using a bibliographic reference manager. The content was summarized by defining all the information to be extracted, such as: author, year, country, journal, sample, objective, results and level of evidence [8] . In the fourth stage, the articles were critically read in order to synthesize the available information and the studies were classified according to the level of scientific evidence according to the Oxford Centre for Evidence-Based Medicine (2009) [10] (Table 2).
Table 1. Search strategies in portals and databases. Curitiba, PR, Brazil, 2023.
Source: The authors (2023).
Table 2. Levels of evidence by type of study. Curitiba, PR, Brazil, 2023.
Source: Adapted from the Oxford Center Evidence-Based Medicine classification (2009).
In the fifth stage, the results were read in full, and analyzed in accordance with the research question. In the sixth stage, the synthesis of the articles and the final considerations of this integrative review were carried out.
3. Results
The search for data resulted in 1808 studies: 705 were excluded for being duplicates and 1103 were selected to read their titles and abstracts. Of these, 794 articles were excluded by reading their titles and 224 after reading their abstracts, resulting in the selection of 85 for full-reading. After this stage, 66 were excluded, making 19 with the addition of another 15 studies, which were identified after reviewing the bibliographic references of the articles selected for the review, thus resulting in the inclusion of 34 studies.
To minimize possible risks of bias in selection of the studies, the stages of reading titles and abstracts, as well as the identification of studies by reading the references, were carried out by two independent evaluators seeking 100% agreement, with a third reviewer evaluating the differences that arose. Figure 1 shows the flowchart corresponding to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) method used to illustrate the selection of articles [11] .
Table 3 shows the categorization of the articles that made up the integrative review corpus by author/year of publication, country of origin, publication journal, sample size, objective, results and level of evidence.
Figure 1. PRISMA flowchart corresponding to selection of the studies. Curitiba, PR, Brazil, 2023. Source: Prepared by the authors (2023), adapted from Page et al., 2021.
Table 3. Characteristics of the studies that made up the integrative review corpus. Curitiba, PR, Brazil, 2023.
ILPI: Instituição de Longa Permanência para Idosos (Long-stay Institution for Older Adults); NGT: Nasogastric Tube; PEG: Percutaneous Endoscopic Gastrostomy; MNA: Mini Nutritional Assessment BMI: Body Mass Index; AO: Aspiration Pneumonia; Kcal: kilocalorie; NOR: Nothing via Oral Route; EOI: Early Oral Intake; HGS: Hand Grip Strength; ICU: Intensive Care Unit; ADLs: Activities of Daily Living; EN: Enteral Nutrition; TPN: Total Parenteral Nutrition; EEN: Early Enteral Nutrition; CMF: Careful Manual Feeding. Source: The authors (2023).
Among the 34 studies analyzed, there was predominance of publications from 2022 (n = 5; 14.70%), followed by 2017 and 2020 (n = 4; 11.76% each year), 2016 and 2019 (n = 3; 8.82% each year), then by 2012, 2014, 2015 and 2018 (n = 2; 5.88% each year), and 2001, 2003, 2005, 2008, 2011, 2013 and 2021 in a smaller percentage (n = 1; 2.94% each year). The materials were published in 28 different journals, with predominance in the Journal of Nutrition Health Aging and Clin Nutr (n = 3; 8.82%).
The following stand out among the countries in which the studies were conducted: Japan (n = 8; 23.53%), China (n = 7; 20.59%), Spain, United States of America (USA) and Israel (n = 3; 8.82% each) followed by Australia and Brazil (n = 2; 5.88% each) and Austria, Italy, France, England, Norway and Vietnam (n = 1; 2.94% each). Regarding the language of the articles, the majority were published in English (n = 32; 94.11%), followed by Spanish (n = 2; 5.89%).
Several studies included the assessment of nutritional intake and/or nutritional status of hospitalized aged people as a central topic [16] as well as nutritional assessment and risk factors for malnutrition [14] , assessment of nutritional intake in longer-lived older adults [40] , assessment of nutritional status and use of handgrip strength as a marker of nutritional risk [27] , in addition to assessing the effects of poor appetite on the health outcomes [26] .
It was observed that the nutritional condition of hospitalized aged people can be influenced by clinical conditions such as advanced dementia [12] [21] [45] , Alzheimer’s dementia with pneumonic symptoms [18] , pneumonia [19] [25] , gastrointestinal tumors [44] , hip fracture [20] [36] , psychiatric disorder [30] , COVID-19 [39] , subacute stroke and oral and intestinal microbiome [34] .
According to the Oxford scale, there was predominance of level of scientific evidence 2B in the studies (n = 26, 76.47%), followed by 1A, 1B and 2C (n = 2; 5.88%) and by 1C and 4 (n = 1, 2.94).
4. Discussion
The main problem listed in relation to the dietary issues was dysphagia, with studies involving clinical assessment [29] , dysphagia in sarcopenics [17] , effects of rehabilitation in dysphagic patients [25] , effects of dysphagia rehabilitation in infectious conditions [22] , assessment of dysphagia and nutritional status [38] , assessment of risk factors for death in the hospital environment [23] , assessment of the association of dysphagia, oral status and malnutrition [24] and assessment and risk factors for dysphagia in burns [31] .
Nutritional issues involving more than one clinical condition such as dysphagia and burns [31] or clinical conditions with nutritional interventions such as assessment of burns and early introduction of dietary supplementation [37] and acute stroke with nutritional interventions [42] [43] were also highlighted.
The assessment of nutritional status, particularly the importance of nutritional level and risk, for hospitalized aged people, was highlighted as predictors of unfavorable outcomes during hospitalization [16] . Malnourished patients and those at risk of malnutrition present lower survival rates (p = 0.001). Lower MNA scores were associated with laboratory markers of malnutrition and were lower in patients with infections, malignancy, pressure injuries, dementia, recent orthopedic surgery and stroke [14] . The nutritional risk factors in older adults are age (p < 0.001), living conditions (p = 0.009), number of CNCDs (p < 0.001), nutritional knowledge (p = 0.007) and difficulties in ADLs (p < 0.001) [40] .
Nutritional status can be assessed by handgrip strength in hospitalized patients [27] . Nutritional support guided by repeated energy balance measurements leads to better outcomes after hip fracture surgery, as there is a lower cumulative energy balance (−1229.9 ± 1763 versus −4975.5 ± 4368 kcal, p = 0.001) and a lower total complication rate (27.3% versus 64.3%, p = 0.012) [20] . The patients subjected to hip arthroplasty who had hypoalbuminemia and received dietary supplements had less wound effusion (OR: 0.57; 95% CI: 0.36 - 0.91, p < 0.05), a lower rate of surgical site infections (OR: 0.40; 95% CI: 0.17 - 0.91, p < 0.05), fewer periprosthetic joint infections (OR: 0.26; 95% CI: 0.08 - 0.79, p < 0.05) and fewer 30-day readmissions (OR: 0.22; 95% CI: 0.06 - 0.79, p < 0.05) [36] .
The evaluation of the nutritional supplementation effects has been the subject of several studies, with different clinical profiles and methodological designs. The nutritional supplementation effects were evaluated in burn patients [37] , in feedback protocols [41] [43] , in reducing the enteral diet administration time in aged people with acute stroke [42] , in aged patients in general [35] , in malnourished patients [15] , in aged people with gastrointestinal tumors [44] , in older adults aged at least 80 years old [40] , in hip surgery patients [20] and in cases of hip arthroplasty with hypoalbuminemia [36] .
Inadequate energy intake is associated with lack of appetite (OR: 1.85; 95% CI: 1.42 - 2.06), high BMI (OR: 1.70; 95% CI: 1.40 - 1.81), diagnosis of infection (OR: 1.70; 95% CI: 1.14 - 1.94), malignancy (OR: 1.79; 95% CI: 1.06 - 2.00), delirium (OR: 1.62; 95% CI: 1.01 - 1.74) and need for assistance with feeding (OR: 1.45; 95% CI: 0.95 - 1.72) [16] . In turn, lack of appetite was associated with increased risk of in-hospital infections (OR: 3.53; 95% CI: 1.48 - 8.41, p = 0.004) and increased risk of death at six months (HR: 2.29; 95% CI: 1.12 - 4.68, p = 0.023) [26] .
Nutritional support reduces hospitalization time and the incidence of infectious complications in patients at nutritional risk [35] and increases the possibility of hospital discharge [15] . Rapid enteral nutrition administration can be used safely and has the potential to decrease the time required for feeding in acute stroke cases with severe dysphagia [42] . Hypocaloric enteral nutrition presented higher mortality at 90 days when compared to modified complete enteral nutrition (with prokinetic agents), with 34% versus 17% (OR: 2.89; 95% CI: 1.46 - 5.72; p = 0.0023), whereas the difference was less significant between hypocaloric enteral nutrition and complete enteral nutrition: 23% (OR: 1.92; 95% CI: 1.00 - 3.69; p = 0.049) [43] . Enteral nutrition support can improve nutritional status, even in the perioperative period of aged patients with gastrointestinal tumors, improving immune function and promoting intestinal peristalsis [44] .
Half of the hospitalized older adults aged at least 80 years old are at nutritional risk. Mortality was associated with age ≥ 85 years old, previous diagnosis of dysphagia, having an oncological disease, comorbidities, low functionality prior to hospitalization, MNA < 17 points, albumin < 3 g/dl, and treatment with opioids, neuroleptics or early antidepressant [23] .
Difficulty eating is an important factor responsible for poor nutrition in hospitalized aged people. In hospitalized aged people affected by dementia, especially in the advanced stage, with Functional Assessment Staging (FAST) of at least 7A, 46.3% use the alternative feeding route. Three-month mortality in such cases was 11.1% in the oral feeding group and 41.9% in the alternative feeding group (p = 0.004), increasing to 27.8% and 58.1% at 6 months (p = 0.012). Higher incidence of Aspiration Pneumonia (AP) was observed in the alternative feeding group (p = 0.006), although there was no difference in the number of hospitalizations between the groups (p = 0.365) [21] .
Patients with recurrent aspiration pneumonia take thickeners more frequently (61.8% versus 11.6%, p = 0.0001). Mortality in these patients was associated with increased age, worse functional capacity and laboratory data indicating malnutrition. Lower albumin levels were associated with higher six-month mortality (OR: 1.13; 95% CI: 1.01 - 1.26, p = 0.03) [18] .
The pneumonia rate is higher in dysphagic patients. Oral intake of a modified texture diet should be a prioritized method, rather than tube feeding or soft/regular food options for dysphagic patients [38] . Patients undergoing dysphagia rehabilitation had a higher proportion of oral intake at discharge (OR: 1.32; p < 0.001). Patients with mild pneumonia had a higher Odds Ratio for total oral intake associated with dysphagia rehabilitation than those with moderate and severe pneumonia (OR: 2.27; p < 0.001) [22] . Dysphagia assessment upon admission can help define more appropriate care [23] .
The main disorder listed in relation to the factors that affect feeding was dysphagia [29] . There is a strong association between dysphagia and malnutrition, multimorbidities, cognitive impairment, greater dependence on ADLs and greater need for care [32] . There was a relationship between dysphagia and salivary hypofunction (p < 0.001), loss of posterior occlusive pairs (p = 0.014), dependence for oral self-care (p < 0.001) and dependence for self-feeding (p < 0.001) [24] . Oral food intake increases diversity and composition of the oral and intestinal microbiome [34] . Swallowing impairment was related to thinness and not to general frailty or presence of dementia [17] .
Dysphagia in severe burns occurs in half of the hospitalized aged people. Dysphagic patients had longer duration of feeding (9.32 days versus 0, p < 0.001), longer hospitalization times (33.61 versus 13.86 days, p < 0.001), more severe in-hospital complications (68% versus 9%, p < 0.001), higher mortality (23% versus 0%, p = 0.003) and greater need for mechanical ventilation (26% versus 0%, p = 0.001). Dysphagia was associated with burn size (OR: 1.24; 95% CI: 1.11 - 1.43, p = 0.0011), prior cognitive impairment (OR: 5.08; 95% CI: 1.37 - 24.68, p = 0.0232) and severe malnutrition (OR: 4.18; 95% CI: 1.27 - 15.62, p = 0.0233) (31). Introducing early diet in burn patients reduces the risks of sepsis by 27% (OR: 0.73; 95% CI: 0.64 - 0.79, p = 0.03), of superficial skin infection by 21% (OR: 0.79; 95% CI: 0.61 - 0.84, p = 0.01), of pneumonia by 15% (OR: 0.85; 95% CI: 0.77 - 0.92, p = 0.01 and of death by 18% (OR: 0.82; 95% CI: 0.66 - 0.98, p = 0.04) [37] .
The predictors for feeding tube placement included African-American ethnicity (OR: 9.43; 95% CI: 2.1 - 43.2) and living in an ILPI (OR: 4.9; 95% CI: 1.02 - 2.50). The infection at admission diagnosis is associated with higher mortality (OR: 1.9; 95% CI: 1.01 - 3.60) [12] . Using a Nasogastric Tube (NGT) was a risk factor for pneumonia (OR: 1.41; 95% CI: 1.08 - 1.85) [45] and for unexpected death due to suffocation (OR: 9.53; 95% CI: 1.03 - 88.26, p = 0.047) [19] . Patients who did not receive an oral diet for up to 48 hours and/or did not have a speech therapy evaluation had lower nutritional intake during hospitalization (p < 0.05) and longer duration of treatment (13 days; 95% CI: 12.04 - 13.96) [25] .
Percutaneous endoscopic gastrostomy was the topic of studies that included the profile of the patients subjected to this procedure [33] , the risk factors that influence regret to perform the procedure [28] and the evaluation of the patients’ microbiota when subjected to different alternative feeding routes [13] .
The main indication for percutaneous endoscopic gastrostomy was dementia, followed by stroke. Malnutrition was observed in 68% at the time of indication. Mortality of the severely malnourished group was 26.4% and 9.5% without malnutrition, with suspicion or malnutrition, p = 0.04 [33] . Even more unfavorable to an adequate diet for hospitalized aged people was pathogenic colonization of the oropharynx in patients on tube-enteral feeding [13] .
This integrative literature review presents the following strengths: the comprehensive search strategy; evaluation of the level of evidence of the studies included; and the standardized data extraction process. As limitations, it can be mentioned that more than half of the studies included in this review were published more than five years ago.
5. Conclusions
The integrative review showed an association between inadequate energy intake and lack of appetite, infections, malignancy, delirium and need for feeding assistance. Assessments of nutritional status and risk factors for hospitalized aged people are fundamental in determining the diet. Nutritional support improved the perioperative nutritional status, in addition to reducing the hospitalization times and the number of infectious complications.
Dysphagia affected nutritional status and was associated with multimorbidities, cognitive impairment, malnutrition, higher dependence for the activities of daily living, and greater care need. The factors evidenced provide geriatric and gerontological knowledge in the face of inadequate intake in hospitalized aged people and guidance for preventive and intervention treatments and care during hospitalization.