Malnutrition among patients in nursing homes and its association with dementia

Abstract

Aim: To reveal the prevalence of malnutrition among patients in nursing facilities using the screening tools Mini Nutritional Assessment (MNA), Malnutrition Universal Screening Tool (MUST), and Nutritional Journal (NJ). The agreement between the tools, and the correlation between malnutrition and severity of dementia, were explored.

Methods: 97 patients living in nursing facilities were included. The patients’ nutritional status was assessed using MNA, MUST, and NJ. Severity of dementia was determined using Clinical Dementia Rating Scale (CDR).

Result: When classifying patients’ nutritional status into two categories (normal vs. risk), the prevalence of risk for malnutrition was 68%, 28% and 35% using MNA, MUST and NJ respectively. Using three categories, the prevalence of high risk for malnutrition was 13%, 11% and 16% using MNA, MUST, and NJ respectively. There was a positive association between dementia severity and worse nutritional status using MNA (p<0.001). This association was not significant using NJ (p=0.223), nor MUST (p=0.303).

Conclusion: The prevalence of malnutrition among patients living in nursing facilities varies according to screening tool applied, and how the result is presented. Risk of malnutrition increases parallel to the severity of dementia regardless of the screening tool, but intensity is more apparent with MNA.

Introduction

Malnutrition or undernutrition (1) refers to a lack of intake or uptake of nutrition leading to altered body composition and body cell mass. Its consequence is diminished physical and mental function, and impaired clinical outcome from diseases (2). Older people are prone to malnutrition due to both biological aging (3), and to chronic conditions associated with malnutrition such as lung and heart diseases, gastrointestinal disorders, dental and oral problems, polypharmacy, depression, and dementia (4). Malnutrition has always been the focus among old patients, although obesity also occurs in this age group (5). Overweight and obesity are defined as abnormal or excessive fat accumulation that may impair health (2).

People with dementia are particularly vulnerable to malnutrition due to a variety of reasons such as difficulty in procurement and preparation of food, feeding difficulties, increased risk of infections (6), and changes in appetite (7). Previous studies have found that malnutrition is associated with dementia severity (8). However,there is also evidence that obesity exists among patients with dementia (9). Today, more than 70 000 people in Norway are living with dementia (10). Most of these patients are old and suffer from multiple diseases. In nursing homes, about 80% of the patients suffer from dementia (11).

The Norwegian Directorate of Health, through the National Professional Guidelines for Prevention and Treatment of Malnutrition, recommends that all patients admitted to health facilities, should undergo nutritional screening upon admission and regularly thereafter (12). The guideline recommends several nutritional screening tools. For older patients, the tools Mini Nutritional Assessment (MNA) and the Malnutrition Universal Screening Tool (MUST) are recommended. The Nutritional Journal (NJ)# Ernæringsjournal, Subjective Global Assessment (SGA), and Nutritional Risk Screening (NRS-2002) are recommended as alternative tools (12, 13).

Despite the recognized importance of malnutrition among patients living in nursing institutions, only a limited number of studies present updated statistics on nutritional status among patients with dementia in Norwegian nursing facilities. Previously presented prevalence estimates of malnutrition in this setting vary from 33 to 69% depending on nutritional screening tool used and group studied (14-16).

Accordingly, the aim of our study was to present the prevalence of malnutrition among patients living in nursing facilities using the three commonly used nutritional screening tools MNA, MUST, and NJ. We also explored the agreement between these three instruments, as well as whether the prevalence of malnutrition varied with severity of dementia.

Figure 1. Patient flow illustrating patient inclusion CRPD (Communal residences for people with dementia)

Methods

Population / Sample

Using a cross-sectional design, we included patients living in nursing homes and communal residences for people with dementia, collectively termed nursing facilities. Inclusion was conducted from December 2016 to March 2017, and took place in the two adjacent municipalities of Sandefjord and Larvik, Norway.

Inclusion criteria were permanent residence or three months continuous residence in one or more nursing facilities. Patients were included regardless of age.

Patients with short-term residence, patients who were acutely ill, patients who did not cooperate when measuring height and weight, and patients who had missing height measurement, were excluded. The sample size was determined by practical considerations.

Data collection

Nurses responsible for the regular care of the patients performed all measurements in which direct contact with the patients were necessary. All data were then collected by the primary investigator (RM) through a series of interviews with the nurses who performed the measurements. The registration forms, MNA, MUST, NJ, Clinical Dementia Rating Scale (CDR), and demographic data were completed during the interviews.

Demographic data

The patients’ age, gender, and length of stay in the institutions, were registered. To avoid direct identification of the patients, their exact age was not recorded. Age was divided in eight categories: less than 65 years, 65-70 years, 71-75 years, 76-80 years, 81-85 years, 86-90 years, 91-95 years, and more than 95 years.

Screening of Nutritional Status

To evaluate the nutritional status of the participants comprehensively, three nutritional screening forms were utilized: MNA, MUST, and NJ. Weight, height, and circumference of arm and leg, measured within the last four weeks before the registration was used. Weight was measured using weighing chair. Height was measured with the patient in standing or lying position. Height measurement from medical record was used if height measurement was difficult.

The original form of the MNA with 18 parameters was used in this study. MNA assesses patients’ weight changes over time, body mass index (BMI) measured askg/m2, mobility, and possible neuropsychological disorders. It also provides information about the patient’s living situation, number of regular medications, presence of wounds or skin sores, number of meals daily, daily consumption of nutrients and fluids, independence during mealtime, and a measurement of mid-arm and calf circumference. The maximum score is 30, whereas a score less than 17 indicate malnutrition, 17 to 23,5 indicate a risk of malnutrition, and a score from 24 to 30 indicate normal nutritional status. MNA also includes two questions were the patients are asked about their own health status compared to others, and their self-view of nutritional status (17). Regarding these two questions the nurses, after discussing with other caregivers, had to answer for those patients unable to speak for themselves due to lack of language or severe dementia.

MUST is a screening tool primarily developed to identify malnutrition in acute settings, and to help draw up an action plan for persons who are undernourished (18). The screening follows three steps where patients' BMI, weight loss over the last three to six months, and possible acute illness is registered. Each item is given a score of 0 to 2, and the risk of malnutrition is given from the sum of these scores. A total added score of 0 indicates low risk, 1 indicates medium risk, and 2 indicate high risk.

NJ identifies the patient’s BMI and weight changes over time, as well as other nutritionally related data such as decrease in appetite, dental problems, chewing or swallowing problems, sore or dry mouth, nausea and vomiting, diarrhoea or constipation, oedema, grabbing or movement problem, independence during mealtime, and vision. The assessment is graded as good nutritional status, risk of malnutrition, and severe malnutrition (19)

Dementia Diagnosis and Severity

Dementia diagnoses registered in the patients’ medical records were used. No direct cognitive assessment or diagnostic process was done during the registration.

Severity of dementia was determined using the CDR (20). The CDR assesses the patients’ memory, orientation, judgment, social activity, home and leisure interests, and ability to care for oneself. The patients were categorized using the following scale: 0 = no dementia, 0.5 = uncertain, 1 = mild, 2 = moderate, and 3 = severe dementia.

Ethical considerations

The project wassubmitted to the Norwegian Center for Research Data (NSD). Informed consent was not required as mapping of nutritional status is included in the routine care in Norwegian nursing homes. All data was registered anonymously as required by the NSD. Health personnel working in the nursing facilities carried out all anthropometric measurements. Exact age, name, and date of birth were not given. The researcher did not have access to patients’ records nor have direct contact with the patients during data entry.

Table 1. Description of the participants, n=97.

Age in years

n (%)

< 65

2 (2.1)

65-70

0 (0.0)

71-75

8 (8.2)

76-80

9 (9.3)

81-85

19 (19.6)

86-90

31 (32.0)

90-95

22 (22.7)

>95,

6 (6.2)

Female, Number of months in institution, mean (SD)*

77 (79.4) 26.0 (20.2)

Body Mass Index (kg/m2)

<20

21 (21.6)

20-24.9

29 (29.9)

25-29.9

31 (32.0)

30

16 (16.5)

Dementia Diagnosis

Alzheimer’s Disease

18 (18.6)

Vascular dementia

5 (5.2)

Lewy Body dementia

1 (1.0)

Mixed Type dementia

4 (4.1)

Other Types of dementia

1 (1.0)

Dementia, no subtype

45 (46.4)

No dementia diagnosis

23 (23.7)

Dementia Severity Using CDR

No dementia

14 (14.4)

Mild Dementia

20 (20.6)

Moderate Dementia

32 (33.0)

Severe Dementia

31 (32.0)

*Number of months are given as mean (SD)

SD=standard deviation

CDR=Clinical Dementia Rating Scale

Statistical Analysis

All statistical analysis was performed using IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY: IBM Corp.

Descriptive analyses of the participants were done initially.  By visual inspection of a histogram and a normal Q-Q plot, all continuous data were normally distributed and therefore presented as mean ± SD.

MNA, MUST, and NJ classify the patients into three risk categories that we labelled low risk (MNA= normal, NJ= good, MUST= low risk), medium risk (MNA= at risk, NJ= risk, MUST= medium risk), and high risk (MNA= malnourished, NJ= severe malnutrition, MUST= high risk).

Nutritional status was then reorganized into two categories to conform with earlier studies comparing screening tools (21, 22): good nutritional status (MNA= normal , Nutritional Journal= good, MUST= low risk), and at risk for undernutrition (MNA= malnourished and at risk, Nutritional Journal= risk and severe malnutrition, MUST= medium risk and high risk). The two categories were used when calculating the variation in nutritional status related to severity of dementia, as well as presenting the agreement between the tools.

Kappa statistics (23) was used to calculate the pairwise agreement between the screening tools.

To explore the association between severity of dementia and nutritional status, the Mantel-Haenszel test was used to calculate linear-by-linear association.

Table 2: Classification of nutritional status using MNA, MUST, and NJ. Values refer to number of participants.

Nutritional Tools

MNA

MUST

Nutritional Status

Good

At risk

Good

At risk

MUST

Good

28

42

N/A

N/A

At risk

3

24

N/A

N/A

NJ

Good

26

37

59

4

At risk

5

29

11

23

MNA= Mini Nutritional Assessment

MUST= Malnutrition Universal Screening Tool

NJ= Nutritional Journal

Agreement between MNA and MUST: Kappa score: 0.20

Agreement between MNA and NJ: Kappa Score: 0.22

Agreement between MUST and NJ: Kappa Score: 0.64

Results

Selection and Description of participants

The patient flow is presented in Figure 1. Among the 114 patients available for inclusion, 73 patients were residents of communal residences for people with dementia, and 41 patients were living in ordinary nursing homes. 14 patients were excluded. Accordingly, the total number of included patients was 97, which was 83% of the initially selected participants.

Among the 97 included patients, 79 were female and 20 were male. The length of stay in institution ranged from three to 74 months. The mean (SD) number of months in institution was 26 (20). A further description of included patients is given in Table 1.

74 of the 97 patients had a formal diagnosis of dementia, and the subtype of dementia was diagnosed in 29 of these patients. Among those with known dementia subtype, 18 had Alzheimer’s dementia, five had vascular dementia, one had Lewy body dementia, four had mixed dementia, and one patient had other type of dementia. The number of patients registered with no dementia diagnosis based on medical record was 23. Using CDR however, 14 patients were registered to have no dementia. According to CDR, 20 of the patients had mild dementia, 32 had moderate dementia, and 31 patients suffered from severe dementia.

Nutritional status using the MNA, the MUST, and the Nutritional Journal

The mean (standard deviation) BMI-score in our patients was 24.6 (4.7). Grouped BMI-scores are presented in Table 1.

Online supplemental figure 1 illustrates the nutritional status of patients with obesity. All 16 patients with obesity had low nutritional risk using NJ. Using the MUST, 15 were at low risk, while one patient was at medium nutritional risk. Using the MNA, five patients were at low risk, 10 at medium risk, and one patient was classified with high nutritional risk.

Online supplemental figure 1 also illustrates the different results MNA, MUST, and NJ yielded using three categories. The percentage of patients in high risk varied from 11.3% using the MUST, 13.4% using the MNA, to 15.5% using NJ.Medium nutritional risk varied from 17.5% using the MUST, 19.6% using the NJ, to 54.6% using the MNA. Low risk varied from 32.0% using the MNA, 64.9% using the NJ, to 72.2% using the MUST. The results of nutritional assessments when reorganized into two categories: at risk for undernutrition and good nutritional status, are presented in Figure 2. The prevalence of at risk for undernutrition was then 68%, 28%, and 35% using MNA, MUST, and NJ respectively. Good nutritional status was found in 32%, 72%, and 65% of the patients using MNA, MUST, and NJ respectively.

Table 3. Patients’ nutritional status related to severity of dementia using Clinical Dementia Rating Scale (CDR) and use of nutritional screening tools.

No dementia

Mild dementia

Moderate dementia

Severe dementia

p-value

NJ

Risk of undernutrition

4

6

10

14

Good nutritional status

10

14

22

17

Mantel-Haenszel-test

0.223

MNA

Risk of undernutrition

2

11

23

30

Good nutritional status

12

9

9

1

Mantel-Haenszel-test

0.001

MUST

Risk of undernutrition

3

5

8

11

Good nutritional status

11

15

24

20

Mantel-Haenszel-test

0.303

MNA= Mini Nutritional Assessment

MUST= Malnutrition Universal Screening Tool

NJ= Nutritional Journal

Mantel-Haenszel test was used to calculate linear-by-linear association, to explore the association between severity of dementia and nutritional status

3.3. Agreement in diagnosing malnutrition between MNA, MUST, and NJ

The agreement between the three screening tools is illustrated in Table 2. MNA and MUST yielded a Kappa score of 0.20. Between MNA and NJ, the Kappa score was 0.22, while the Kappa score between MUST and NJ was 0.64.

Nutritional status related to severity of dementia

The relationship between nutritional status using NJ, MUST and MNA, and severity of dementia using the CDR is illustrated in Figure 3 (available online). The number of patients in each group in Figure 3 is given in Table 3.

Using the Mantel-Haenszel test, there was a positive correlation between the severity of dementia and the risk for undernutrition using MNA (p<0.001). This was not observed for severity of dementia and risk of undernutrition was not found using MUST (p= 0.303), nor NJ (p= 0.223).

Discussion

Our study of 97 patients living in nursing facilities shows that the prevalence of undernutrition varies greatly according to screening tools used and the way the outcome is presented. When presenting the result of the nutritional screenings using three categories, the MNA classified 54.6% of the patients in medium nutritional risk, compared to 17.5% and 19.6% for MUST and NJ respectively. Although the result varied minimally at high risk, which was 13.4%, 11.3% and 15.5% using MNA, MUST and NJ respectively, the outcome changed drastically when the results were organized into two categories to conform with earlier studies (21). With two categories, the prevalence of risk for undernutrition was 68%, 29%, and 35% using MNA, MUST and NJ respectively. Velasco and colleagues (21), presented their results using two categories, and got the same result as our study: the MNA presents a much higher prevalence of undernutrition than the MUST. We found that it was the patients in medium risk for undernutrition who mainly accounted for the difference between the screening tools. Moreover, some recent local studies using either MNA or MUST (14-16) found comparable prevalence as our study using three categories, and when transforming their results into two categories, MNA would present a higher prevalence of risk of undernutrition than MUST.

The agreement between the tools calculated using Kappa-statistics, revealed that the different tools did not necessarily identify the same patients at risk of undernutrition. MUST and NJ showed best agreement, while MNA had slight agreement with the other tools.

Likewise, obese patients were placed in different categories depending on the tool used. That patients are classified differently depending on which instrument used is not surprising as the majority of the tools’ questions are not identical. Also, in shared questions, different cut-offs are applied. BMI for instance is common for the three instruments, but each uses different BMI cut-offs as an indicator of undernutrition.

Several studies have been conducted to compare the validity and reliability of the different nutritional screening tools (21, 24, 25). MNA is the most well validated tool for older persons (26) in hospitals and private homes (27) . It is also validated for older persons in long-term care (28). Likewise, MNA is used as a gold standard along with SGA when comparing other nutritional screening tools for validity and accuracy (24) .MUST is alsoa well validated tool (29, 30) and has been used in several comparative studies; although it was not specifically validated for older persons in long term institutions.Among the three screening tools used in our study, MUST has the lowest BMI cut-off with a BMI of 20 or higher considered as normal. As a BMI above 23.0 is recommended for older people (31) due to reduced mortality, the use of MUST may lead to underdiagnosis, unless the BMI cut-off is adjusted.

NJ is a locally developed tool, which is widely utilized in many Norwegian municipalities. However, it is to our knowledge not validated (13) , and has not been included in comparative studies before.

In their guideline for prevention and treatment of malnutrition, the Norwegian Directorate of Health does not comment on validation of the tools it recommends, nor reveals that the different tools may produce quite different outcomes. Various municipalities choose different tools due to lack of a gold standard. When different nursing facilities in one municipality use different tools, their statistic will not be comparable. Furthermore, since the use of different tools could lead to different diagnosis, some patients might not get the right treatment nor covered their nutritional needs, which can have major consequences for the patient`s health.

A recent study concluded that the degree of malnutrition is correlated with the severity of dementia irrespective of the type of dementia (8). This means that as dementia progresses, the risk for undernutrition increases. Our study has a similar result, but the degree of correlation depends on nutritional screening tool used. With the use of NJ and MUST, the risk for undernutrition increases at a gradually low pace from no dementia to severe dementia. With the use of MNA though, the rate increases at a regular high pace from no dementia to severe dementia. Presence of dementia is one of the components in MNA, which may contribute to its association with severity of dementia. Whether this may lead to an over diagnosis of nutritional risk in patients with dementia is not known.

Figure 2: The patients’ nutritional status using MNA (Mini Nutritional Assessment), MUST (Malnutrition Universal Screening Tool), and NJ (Nutritional Journal), categorized into two nutritional status. N=97. The values reflect prevalence of risk for malnutrition, and good nutritional status.

Strengths and limitations of the study

Strengths of our study is that the same interviewer conducted all the interviews with care-personnel. Only care-personnel with good knowledge of the included patients’ health were interviewed. Moreover, except from the NJ, only validated instruments measuring both degree of dementia and nutritional status were used.

Some limitations must also be addressed. Old height measurements were used if difficult to measure accurate height. In addition, some of the patients with dementia were unable to give self-report of own health- and nutritional status. Another limitation is the lack of direct assessment of cognitive function, and that the severity of dementia was based on CDR alone. The sample size is also small, and a larger study may yield more exact data.

The aim of our study was not to reveal what tool is most suitable for older patients in nursing facilities, nor to discuss the validation or sensitivity issues among the screening tools. However, these important issues need further research.

Implications for Practice

When assessing the nutritional status of older patients in nursing facilities, it is important to be aware that different instruments may yield different results. The same patients’ nutritional status may therefore change if using different screening tools.

Close monitoring of the nutritional status of patients with dementia is especially important, as dementia poses risks to patients` nutritional health. Healthcare professionals should also be aware that as dementia progresses, nutritional problems may also increase. Individual nutritional care plan and management is therefore vital for these patients.

Conclusion

The prevalence of malnutrition among patients living in nursing facilities vary depending on the nutritional screening tools applied and how the result is presented. The risk of undernutrition among patients with dementia increases with the severity of dementia, regardless of screening tool, however, the intensity was most apparent with MNA. Obesity is also found among older patients in the institutions, but they are classified into different nutritional status depending on the screening tool used.

Acknowledgement:

Special thanks to the staff of the following for their participation during the data collection: The communal residences for people with dementia in Sandefjord (Nygårdsvollen, Ranvikskogen & Bøkeveien); Mosserødhjemmet, Post 1, 2 & 3; Tjølling Nursing Home, Huseby Ward & Kaupang Ward.

Funding

No funding was received for the implementaton of the project. This project is a part of a master thesis as a completion of the program Master in Geriatric Health Care. The project was implemented in cooperation with the University of South-Eastern Norway, municipality of Sandefjord and municipality of Larvik.

Conflict of Interest

None.

References

  1. Katsilambros N, Dimosthenopoulos C, Kontogianni M, et al. Clinical Nutrition in Practice. United Kingdom: Blackwell Publishing Ltd; 2010.

  2. Cederholm T, Barazzoni R, Austin P, et al. ESPEN Guidelines on Definitions and Terminology of Clinical Nutrition2016.

  3. St-Onge M-P, Gallagher D. Body composition changes with aging: The cause or the result of alterations in metabolic rate and macronutrient oxidation? Nutrition 2010;26(2):152-5.

  4. Sortland K. Eldre og Ernæring (Older people and nutrition). In: Bondevik M, Nygaard HA, editors. Tverrrfaglig Geriatri (Interdisciplinary geriatrics). 3rd edition ed. Bergen: Bokforlaget; 2012. p. 157- 81.

  5. Mathus-Vliegen EM. Obesity and the elderly. Journal of clinical gastroenterology. 2012;46(7):533-44.

  6. Pivi GAK, Bertolucci PHF, Schultz RR. Nutrition in Severe Dementia. Current Gerontology and Geriatrics Research. 2012;2012:983056

  7. Kai K, Hashimoto M, Amano K, et al. Relationship between eating disturbance and dementia severity in patients with Alzheimer’s disease. PLoS ONE. 2015;10(8)

  8. Camina Martin MA, Barrera Ortega S, Dominguez Rodriguez L, et al. [Presence of malnutrition and risk of malnutrition in institutionalized elderly with dementia according to the type and deterioration stage]. Nutricion hospitalaria. 2012;27(2):434-40.

  9. Bednarska-Makaruk M, Graban A, Wiśniewska A, et al. Association of adiponectin, leptin and resistin with inflammatory markers and obesity in dementia. Biogerontology. 2017;18 (4): 561-580

  10. Strand BH, Tambs K, Engedal K, et al. [How many have dementia in Norway?]. Tidsskrift for den Norske laegeforening. 2014;134(3):276-7

  11. Nazarko L. Maintaining good nutrition in people with dementia. Nursing and Residential Care. 2013;15(9):590-5.

  12. Guttormsen AB, Hensrud A, Irtun Ø, et al. Nasjonale faglige retningslinjer for forebygging og behandling av underernæring (National professional guidelines for prevention and treatment of malnutrition ). Oslo:Helsedirektoratet;2013.

  13. Gjerlaug AK, Harviken G, Uppsata S, et al. Verktøy ved screening av risiko for underernæring hos eldre (Tools for screening the risk of malnutrition in the elderly). Sykepleien Forskning.2016 11(2)(148-156)

  14. Eide HD, Aukner C, Iversen PO. Nutritional status and duration of overnight fast among elderly residents in municipal nursing homes in Oslo. VÅRD I NORDEN 2012;32:20–4.

  15. Hagen K. Uten mat og drikke… Ernæring Kroken sykehjem 2011 (Without food and drink ...Nutrition Kroken Nursing Home 2011). Utviklingssenter for sykehjem i Troms; 2012.

  16. Aukner C, Eide HD, Iversen PO. Nutritional status among older residents with dementia in open versus special care units in municipal nursing homes: an observational study. BMC geriatrics. 2013;13:26.

  17. Veiledning for utfylling av skjema for ernæringsvurdering: Mini Nutritional Assessment (A guide to completing the Mini Nutritional Assessment MNA) [Internet]. Nestlé Group. Available from: http://www.mna-elderly.com/forms/mna_guide_norwegian.pdf.

  18. Elia M, Baxter J, Carole Glencorse, Jackson A, et al. “MUST” Brosjyren (“Malnutrition Universal Screening Tool”). The British Association for Parenteral and Enteral Nutrition. 2003.

  19. Aagaard H, Roel S. Utvikling av ernæringsjournal: beskrivelse av ernæringsjournalen og dens praktiske gjennomføring foretatt av sykepleiestudenter (Development of Nutritional Journal: Description of nutritional journal and its practical implementation). Halden: Høgskolen i Østfold; 2004.

  20. C. P. Hughes, Berg L, L.Danziger W, et al. A new clinical scale for the staging of dementia (abstract). British journal of psychiatry 1982;140(6):566-72.

  21. Velasco C, Garcia E, Rodriguez V, et al. Comparison of four nutritional screening tools to detect nutritional risk in hospitalized patients: a multicentre study. European journal of clinical nutrition. 2011;65(2):269-74.

  22. Gerasimidis K, Drongitis P, Murray L, et al. A local nutritional screening tool compared to malnutrition universal screening tool. European journal of clinical nutrition. 2007;61(7):916-21.

  23. Viera AJ, Garrett JM. Understanding Interobserver Agreement: The Kappa Statistic. Family Medicine. 2005;37(5):360-3

  24. Young AM, Kidston S, Banks MD, et al. Malnutrition screening tools: Comparison against two validated nutrition assessment methods in older medical inpatients. Nutrition.29(1):101-6.

  25. Neelemaat F, Meijers J, Kruizenga H, et al. Comparison of five malnutrition screening tools in one hospital inpatient sample. Journal of Clinical Nursing. 2011;20(15-16):2144-52.

  26. MNA- Mini NUtritional Status: Nestle Nutrition Institution; 2004 [Available from: http://www.mna-elderly.com/validity_in_screening_tools.html.

  27. Vellas B, Guigoz Y, Garry PJ, et al. The Mini Nutritional Assessment (MNA) and its use in grading the nutritional state of elderly patients. Nutrition. 1999;15(2):116-22.

  28. Christensson L, Unosson M, Ek AC. Evaluation of nutritional assessment techniques in elderly people newly admitted to municipal care. European journal of clinical nutrition. 2002;56(9):810-8.

  29. Skipper A, Ferguson M, Thompson K, et al. Nutrition Screening Tools. Journal of Parenteral and Enteral Nutrition. 2011;36(3):292-8.

  30. Validated Malnutrition Screening and Assessment Tools: Comparison Guide In: (NEMO) NEMO, editor.: the State of Queensland (Queensland Health); 2014.

  31. Mowe M. [Treatment of malnutrition in elderly patients]. Journal of the Norwegian Medical Association (Tidsskr Nor Laegeforeng). 2002;122(8):815-8.