Evaluating the effectiveness of five screening tools used to identify malnutrition risk in hospitalized elderly: A systematic review

Geriatric Nursing xx (2017) 1e8

Contents lists available at ScienceDirect

Geriatric Nursing journal homepage: www.gnjournal.com

Feature Article

Evaluating the effectiveness of five screening tools used to identify malnutrition risk in hospitalized elderly: A systematic review Brooke L. Cascio, MS, RDN, LD a, *, John V. Logomarsino, PhD, RD, LD/N b a b

Orlando Health and Rehabilitation Center, 830 W 29th Street, Orlando, FL 32805, USA Central Michigan University, Department of Human Environmental Studies, 842 Maybank Loop, The Villages, FL 32162, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 15 January 2017 Received in revised form 9 July 2017 Accepted 17 July 2017 Available online xxx

This systematic review investigated 5 frequently used nutrition screening tools (NSTs) used in hospitals and their effectiveness at identifying malnutrition risk in the elderly. A literature review was conducted to obtain research articles focused on malnutrition screening in hospitalized elderly and effectiveness of the NST used. Twenty six articles were reviewed and evaluated, resulting in 8 that met inclusion criteria. The Mini Nutritional Assessment-Short Form, designed for use in the elderly, resulted in overestimation of malnutrition. Four screening tools did demonstrate more effectiveness in identifying malnutrition risk; however, several different biochemical and anthropometric parameters were used, which prevented meaningful comparisons. There is a need for a universal NST “gold standard” for use in the elderly, and further research is indicated. Ó 2017 Elsevier Inc. All rights reserved.

Keywords: Malnutrition Aged Mass screening Hospitalization Nurses Review Systematic

Introduction Nutrition screening tools (NSTs) have been utilized in the hospital setting to alert nurses and other health care professionals of a potential or actual risk of malnutrition. There is a lack of consensus throughout the nursing community to state with complete confidence that one screening tool performs best with the elderly population.1 Several NSTs have been observed in research studies, which has created more confusion and less certainty about which tool is the most accurate. To date, there has been no gold standard definition for malnutrition, or strict clinical parameters for diagnosis.2 Malnutrition is loosely defined as an undernutrition or a deficiency of nutrition.3 Malnutrition is currently diagnosed through various signs and symptoms, such as weight loss, low serum albumin, infection, and muscle wasting.1 The following malnutrition screening tools are commonly used: Mini Nutritional Assessment-Short Form, Nutritional Risk Screening, Malnutrition Universal Screening Tool, Malnutrition Screening Tool, and Geriatric Nutrition Risk Index. If malnutrition risk is determined in the

* Corresponding author. Orlando Health and Rehabilitation Center, 830 W 29th Street, Orlando, FL 32805, USA. E-mail addresses: [email protected] (B.L. Cascio), [email protected] (J.V. Logomarsino). 0197-4572/$ e see front matter Ó 2017 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.gerinurse.2017.07.006

elderly during the initial screening, a consultation is forwarded to a registered dietitian for a more in-depth assessment. The Mini Nutritional Assessment and Subjective Global Assessment are frequently used when a nutrition risk is identified with a NST. In addition to uncertainty about what screening methods are most effective, a lack of timely screening and misidentification of nutritional status has been an area of concern. Mortality, infection, delayed healing, lengthy hospital stay, and increased healthcare costs are negative factors associated with inaccurate screening.4 It is estimated that costs associated with malnutrition increase hospital expenses by 30e70%.5 Screening within 72 h of admission and accurate risk identification will alert nurses, physicians, and dietitians of high-risk elderly patients. In a Dutch study that investigated the diagnosis of malnutrition by healthcare professionals, 15.3% of patients received a screening within 72 h by a physician, 52.8% by medical students, and 29.9% by nurses.6 In addition, timely and accurate screening will result in a faster assessment and intervention to ensure the best possible chance at health improvement and decreased mortality. This systematic review investigates the 5 commonly-used NSTs in hospitals at identifying malnutrition risk in the elderly. Reliability, validity, specificity and sensitivity are important indicators in research that evaluate screening tools. Validity applies to the design and methods of the research and identifies the accuracy of the research. Reliability evaluates the degree to which the

2

B.L. Cascio, J.V. Logomarsino / Geriatric Nursing xx (2017) 1e8

NST produced a stable result. It is important to state that reliability has to be present in addition to validity. Specificity measures if there is a correct identification of studies with no malnutrition risk. Sensitivity measures if there is a correct identification of studies that have malnutrition risk. NSTs have been designed to alert staff of a potential nutrition risk that requires further assessment. Commonly used NSTs included in this review were Malnutrition Universal Screening Tool, Mini Nutritional Assessment-Short Form, Malnutrition Screening Tool, Geriatric Nutrition Risk Index, and Nutritional Risk Screening. These are summarized in Table 1. There is some confusion with screening and assessment tools being used interchangeably when they are not designed for the same purpose. Mini Nutritional Assessment is lengthy to complete, so time management may be an issue if a nurse has a large nurse-to-patient ratio. Some of the NSTs are easy to administer (Mini Nutritional Assessment-Short Form, Nutritional Risk Screening, and Malnutrition Screening Tool), while other NSTs and assessment tools require a more in-depth clinical observation, calculations and laboratory values (Mini Nutritional Assessment, Subjective Global Assessment, Geriatric Nutrition Risk Index and Malnutrition Universal Screening Tool).7 Initial screening is most often conducted by nurses or physicians during the first 48e72 h of admission. Hospitals use a variety of screening methods, which further complicate proper diagnosis. These tools are sometimes difficult to administer and challenging for the front line staff who are completing the initial screening. One study found that only 4% of patients were screened for malnutrition risk, despite 57% of patients who were, in fact, malnourished.8 This further emphasizes the importance of a streamlined screening process that is quick and accurate for nurses to complete. Methods A systematic review was conducted using Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA)9 guidelines to answer this research question: Are five commonly used screening tools effective at identifying malnutrition risk in the hospitalized elderly?

identifiers for malnutrition risk. These identifiers included height (ht), weight (wt), body mass index (BMI), serum albumin, total lymphocyte count, tricep skinfold thickness, calf circumference, and mid-arm circumference. Other variables monitored to identify malnutrition risk included: length of stay, in-hospital mortality, 3 and 6-month mortality, results of different screening tools, physical assessment, and observation. Exclusion criteria were people below 60 years of age, community dwelling or nursing home residents, non-hospitalized, or screening that was completed beyond 72 h of admission, or use of NSTs not included in this review. Titles and abstracts were reviewed to determine if inclusion criteria and the research question were met. Assessment of quality of research studies obtained in this systematic review Quality assessment was completed by using criteria established by the Downs and Black checklist.10 Twenty seven items were evaluated to determine quality, as summarized in Table 2. “Yes” or “no” questions were included and there was a maximum numeric score of 27 points possible. The Downs and Black quality checklist produced scores ranging from 8 to 20 (summarized in Table 3). Although there was some variation in the studies reviewed, this did not affect interpretation of results because of the heterogeneous nature of the studies. The Downs and Black quality checklist has been shown to be effective in the systematic review process.10 Results The search resulted in 1240 total references: 712 in CINAHL, 325 in Web of Science, 52 in PubMed and 151 in Google Scholar. The literature search was not limited by publication date. Duplicates were removed from the databases and 1140 papers remained. Review of title and abstract resulted in 26 papers that were obtained for full-text review. This resulted in 8 studies, which evaluated five NSTs that met established inclusion criteria (Fig. 1). There were no conflict of interests found in the studies reviewed. Research summaries

Search strategy Several databases were used to search for relevant literature. A literature search of PubMed was performed to identify relevant sources. The following Medical Subject Heading (MeSH) terms were used to identify initial sources: malnutrition, mass screenings, aged, nutrition and hospitalized. Cumulative Index to Nursing and Allied Health Literature (CINAHL) search terms were used to identify relevant sources: malnutrition, screening, elderly and hospitalized. A Web of Science search was conducted using following terms: TS ¼ aged AND TS ¼ malnutrition AND TS ¼ mass screening. A free text search was conducted using the words “malnutrition”, “screening tool” and “elderly” to identify additional sources. An exhaustive literature search was performed using these search tools as well as Google Scholar. References were also located through hand-searching of the bibliographies of research articles. No language restrictions were included in the search criteria. Inclusion criteria for studies used in the systematic review included people who were 60 years of age and above, were hospitalized, had malnutrition, had used one of NSTs: Mini Nutritional Assessment-Short Form, Nutritional Risk Screening, Malnutrition Screening Tool, Geriatric Nutrition Risk Index and Malnutrition Universal Screening Tool within 72 h of admission. Although there is no specific tool that accounts for dementia, people with dementia were included in the review. In many of the articles reviewed, similar anthropometric and biochemical data were used as key

Studies have compared screening methods to determine which is most accurate and gives the best indication of malnutrition. Parameters used to evaluate the effectiveness of NSTs are described in Table 3. A single-center, cross-sectional study in Switzerland compared the Mini Nutrition Assessment to the Nutritional Risk Screening NST13 The purpose was to determine if there was a correlation with protein markers used to identify malnutrition. Biochemical values that were obtained within 24 h of admission included albumin, prealbumin, retinol-binding protein, total lymphocyte count, creatinine and c-reactive protein markers. It is important to note that in 60% of the observations, classifications of Mini Nutrition Assessment and Nutritional Risk Screening scores were not in agreement. Nutritional Risk Screening identified more patients who were at risk or malnourished than did the Mini Nutrition Assessment. Some patients who were classified as severe nutrition risk in Nutritional Risk Screening, were at low or no risk according to Mini Nutrition Assessment.13 Patients who had the highest Mini Nutrition Assessment malnutrition risk scores (23 patients) had only median values of albumin, prealbumin, and retinol binding protein markers. The Mini Nutrition Assessment did not reflect any differences in the total lymphocyte count and c reactive protein markers. Patients who were classified as malnourished using the Nutritional Risk Screening had a lower prealbumin and retinol binding protein. The protein marker that was proven to be effective

Table 1 Screening tools used to identify risk of malnutrition in the elderly. Screening Tool

Goal

MST (Malnutrition Screening Tool)

Identify malnutrition in 2 questions; Recent acute adults inpatients unintentional weight & outpatients, including loss, poor appetite. elderly

Anthropometric assessment

Functional assessment

wt loss

Dietetic assessment

Subjective assessment

Strengths

Weakness

Food intake

None

Validated for use in acute hospital and ambulatory care settings; Easy to administer e only two questions. <5 min to complete; Easy to administer; Designed for use in elderly.

Originally developed for the younger adult population (mean age ¼ 57.7 years).

6 questions; Identical to wt, ht, wt loss, CC. the first 6 questions in full MNA.

Presence of dementia, Food intake, presence None of chewing/swallowing functional decline, mobility, acute disease. difficulties.

5 steps

BMI, wt loss

Acute disease condition.

BMI <20, wt. loss in past 3 months.

Acute illness, age.

For the initial 4 Nutritional Risk Identify hospitalized questions, if “yes” is Screening 2002 (NRS) patients who are malnourished or at risk. answered to any, then final screening is completed. If “no” is answered for all, then no risk of malnutrition is identified. Geriatric Nutrition Risk Used to identify GNRI formula uses ideal wt Lorentz formula Index (GNRI) nutrition risk in (WLo) [1.489  alb (g/ sub-acute setting L)] þ[41.7  (weight/ WLo)]

Age wt loss 5e10%; abnormal albumin e 38, 35 and 30 g/L. Four grades of nutritionrelated risk: major risk (GNRI: 82); moderate risk (GNRI: 82 to 92); low risk (GNRI: 92 to 98); and no risk (GNRI: 98).

If unable to answer, caregiver should be present.

Can be used by all care If BMI or ht/wt cannot be obtained, workers. professional judgement by administrator is used to estimate. Reduced dietary intake None Constructed from well in last week. supported studies; Easy to administer; Accounts for age >70.

Does not ask about food intake; Best used in community setting; Use of subjective judgement by administrator. Not designed specifically for elderly.

None

Uses ideal wts; may not be completely accurate.

None

None

Accurate and simple tool, only requires weight and albumin levels; Good indicator of morbidity and mortality in hospitalized elderly.

B.L. Cascio, J.V. Logomarsino / Geriatric Nursing xx (2017) 1e8

MNA-SF (Mini nutrition Identify malnutrition risk in elderly patients; assessment- short If MNA-SF is positive, form) then full MNA must be completed. Identify adults who are MUST (Malnutrition Universal Screening malnourished, at risk, or who are obese. Tool)

No. of items

BMI: body mass index; MAC: mid arm circumference; CC: calf circumference: TSF: triceps skinfold thickness; wt: weight; n: nausea; v: vomiting.

3

4

B.L. Cascio, J.V. Logomarsino / Geriatric Nursing xx (2017) 1e8

at identifying malnutrition risk with the Nutritional Risk Screening was retinol binding protein.13 Nutritional Risk Screening was thought to be an effective screening tool that allows for a quicker identification, especially in cases of acute co-morbidities; however, its initial design was not for use in the elderly population. The researchers concluded that Nutritional Risk Screening was able to identify patients at risk or had the potential of risk. The retinol binding protein was found to be the only effective protein marker in this study that was positively associated with Nutritional Risk Screening and risk of malnutrition.13 A prospective cohort study was conducted in Egypt to identify the prediction of nutrition-related complications using the Mini Nutrition Assessment and Geriatric Nutrition Risk Index.6 The Geriatric Nutrition Risk Index is a newer NST that was developed with the purpose of identifying malnutrition in the geriatric population. Three objective measurements were used to identify if malnutrition were present: albumin, wt, and ht. BMI, ht, wt, midarm circumference, and calf circumference were evaluated along with total protein, albumin, and prealbumin markers. A 6-month follow-up study was conducted to determine if risk of malnutrition according to Mini Nutrition Assessment was associated with increased length of stay, health complications, and mortality. Geriatric Nutrition Risk Index and Mini Nutrition Assessment were similar in the ability to detect in hospital mortality. However, Geriatric Nutrition Risk Index showed greater ability to predict both 3 and 6 month mortality than Mini Nutrition Assessment.6 The Mini Nutrition Assessment and Geriatric Nutrition Risk Index patients classified as malnourished were associated with an increased risk of complications, including infection. However, the Geriatric Nutrition Risk Index NST has not been well researched in terms of effectiveness in identifying malnutrition risk. A cross-sectional study in Indonesia evaluated the effectiveness of Mini Nutrition Assessment-Short Form against biochemical data albumin, hemoglobin, and total lymphocyte count and anthropometric data BMI and mid-upper arm circumference to identify risk of malnutrition.12 Biochemical data were obtained from the medical records, while anthropometric data were obtained by direct measurement <48 h of admission. Patients identified at risk according to Mini Nutrition Assessment-Short Form were noted to have lower biochemical values, including albumin, hemoglobin, and total lymphocyte count. A Korean study evaluated the Mini Nutrition Assessment, Mini Nutrition Assessment-Short Form, Malnutrition Universal Screening Tool, Nutritional Risk Screening, and Geriatric Nutrition Risk Index to determine which was most appropriate at predicting and identifying malnutrition risk.11 If risk was identified in 4 out of 5 NST and 1 assessment tool, they were classified as malnourished. Specific anthropometric data used were ht, wt, BMI, mid-arm circumference and calf circumference and wt loss in a 90 day duration; biochemical data included albumin and total lymphocyte markers.11 In all of the NSTs, anthropometric and biochemical values decreased as risk of malnutrition increased. Mini Nutrition Assessment-SF overestimated malnutrition while the Malnutrition Universal Screening Tool had the greatest sensitivity and specificity, which indicated that it is the most useful tool.18 However, it is important to note that Malnutrition Universal Screening Tool was designed to identify malnutrition in younger adults in the community setting.2,11,17 A randomized pilot study conducted in the United Kingdom evaluated the accuracy of Malnutrition Universal Screening Tool compared to a dietitian’s assessment.16 Extensive training was provided to staff on administration of Malnutrition Universal Screening Tool, which included one-on-one training and the procedure to obtain anthropometric measurements. An aggressive goal was set to screen patients within 6 h of admission. This procedure

Table 2 Downs and Black criteria checklist to determine study quality. (Each question requires a yes or no answer, 1 point is given for yes for a maximum of 27 points)10 1. Is the hypothesis/aim/objective of study clearly described? 2. Are the main outcomes to be measured clearly described in the Introduction or Methods section? 3. Are the characteristics of the patients included in the study clearly described? 4. Are the interventions of interest clearly described? 5. Are the distributions of principal cofounders in each group of subjects to be compared clearly described? 6. Are the main findings of the study clearly described? 7. Does the study provide estimates of the random variability in the data for the main outcomes? 8. Have all important adverse events that may be a consequence of the intervention been reported? 9. Have the characteristics of patients lost to follow up been described? 10. Have actual probability values been reported for the main outcomes? 11. Were the subjects asked to participate in the study representative of the entire population from which they were recruited? 12. Were those subjects who were prepared to participate representative of the entire population from which they were recruited? 13. Were the staff, places, and facilities where the patients were treated, representative of the treatment the majority of patients receive? 14. Was an attempt made to blind study subjects to the intervention they have received? 15. Was an attempt made to blind those measuring the main outcomes of the intervention? 16. If any of the results of the study were based on “data dredging” was the made clear? 17. In trials and cohort studies, do the analysis adjust for different lengths of follow-up of patients? In case-control studies, is the time period between the intervention and outcome the same for cases and controls? 18. Were the statistical tests used to assess the main outcomes appropriate? 19. Was compliance with the interventions reliable? 20. Were the main outcome measures accurate? 21. Were the patients in different intervention groups? 22. Were study subjects in different intervention groups? 23. Were study subjects randomized to intervention groups? 24. Was the randomized intervention assignment concealed from both patients and health care staff until recruitment was complete and irrevocable? 25. Were they adequate adjustment for cofounding in the analysis form which the main findings were drawn? 26. Were losses of patients to follow-up taken into account? 27. Did the study have sufficient power to detect a clinically important effect?

resulted in missed data and inaccurate results due to time constraints. Researchers reevaluated the ambitious screening goal and increased the screening period to within 12 h of admission, which resulted in improved accuracy.16 The study showed that the Malnutrition Universal Screening Tool could be used as an effective NST to identify patients at malnutrition risk. In a separate study conducted in the United Kingdom, Malnutrition Universal Screening Tool was used to screen 150 elderly patients admitted to the emergency room.17 Predictive validity was accessed using length of hospital stay, discharge mortality, and 3 and 6 month mortality as they related to the Malnutrition Universal Screening Tool classification category of malnutrition. BMI, weight loss and acute disease effect were included to assess malnutrition risk.17 Malnutrition Universal Screening Tool was effective and indicated a lower BMI and weight loss in 3e6 months in patients with malnutrition risk. In-hospital mortality rose as Malnutrition Universal Screening Tool’s classification category of malnutrition worsened. Patients classified as medium or high risk also showed a significant increase in post-discharge mortality at 3 and 6 months. An important factor in this study was that only 84 of 150 patients could be weighed; the wt of the other patients had to be obtained through patient or family recall. Ht was obtainable in only 115 patients. This study demonstrated good validity of the Malnutrition Universal Screening Tool in predicting mortality, length of stay, and clinical outcomes. However, due to the acutely ill elderly subject

Table 3 Parameters used to evaluate effectiveness of nutrition screening tools. Author

Screening assessment/ Screening within 72 h

Design/Sample size/ Duration/Location

Baek et al. 201511

MNA-SF, MNA, MUST and NRS, GNRI; Screening upon admission.

Cross sectional design; 141 subjects; (86 males, 55 females) Ages 60e95. July 1-December 2013. Korea

Meyasari et al. 201412 MNA-SF; <48 h

Drescher et al. 201013 MNA vs NRS; <24 h

Variables compared

Malnutrition risk

Wt, wt. loss, BMI, MAC, CC, Confirmed at risk: 44%. Identification of MUST: 36.2%; MNA-SF malnutrition risk in 4 NST alb, tlc. 72.3%. and 1 assessment tool. Confirmed at no risk: 56%. MUST: 63.8%, MNA-SF: 27.7% MNA-SF high risk BMI, MUAC, nutrition Anthropometric (direct Cross sectional design; indicated lower assessment; measurement); 268 subjects; Ages 60biochemical values. 96(160 male, 108 female). Biochemical data (medical alb, hgb, tlc. p < 0.05. record). FebruaryeMay 2014. Indonesia

Single center cross sectional study; 104 subjects; Median age ¼ 84; 81 females, 23 males. AugusteDecember 2007. Switzerland

Anthropometric; Biochemical.

Anthropometric; NRS-2002, SGA, MNA, MST, Functional status, LOS.

Wu et al. 201215

Cross-sectional study; 204 subjects Mean age ¼ 74.3; 143 women, 64 men. 4 month duration; 2006. Portugal MST compared to SGA; RCT 157 patients; Mean Within 72 h of admission. age ¼ 77.6 years, 121 women, 36 men; September 2008eMarch 2010. New Zealand

Farrer et al. 201316

MUST; <12 h.

Random pilot study; 3 elderly care wards (850beds); 2009e2011. United Kingdom

Anthropometric

Stratton et al. 200617

MUST; <72 h

150 emergency room admissions; 100 female/ 50 male mean age 85 (SD 5.5). United Kingdom

LOS days; In hospital mortality; Post discharge mortality after 3e6 months

NRS-2002, SGA, MNA, MST, anthropometry. <48 h.

Patients screened using MST; Assessment data collected using SGA (medical history, physical exam).

Statistical significance/ Downs and Black quality score

Outcome

MUST se: 80.6%; sp: 98.7%. MNA se:100%; sp: 60.8%. NRS se: 100%; sp: 8.5%. MNA SF-se: 100%; sp: 49.4%. Quality score ¼ 12.

MNA-SF showed overestimation of malnutrition compared to the other tools. MUST demonstrated the greatest validity. MNA-SF inferior; tends to overestimate malnutrition risk

MNA-SF classified 78% at risk of malnutrition compared to 69.03% by. anthropometry/ biochemical parameters. Quality score ¼ 12. MNA ¼ 23% malnourished, NRS was effective in High malnutrition risk CC, MUAC; MNA: alb e 30 mg/l, pab e 48% at risk; NRS ¼ 34% at identifying at- risk elderly alb, pab, rbp, 156 mg/l, rbp e 37 mg/l; moderate/severe risk; 66% during acute illness. lymphocytes, crp. low risk. 60% of NRS high malnutrition observations had risk: alb 31 mg/l; discrepancies. Quality p ¼ 0.45; pab- 138 mg/l, score ¼ 16. p ¼ 0.06; rbp- 34 mg/l, p < 0.001. rbp was the only biochemical value associated with NRS and MNA scores that indicated malnutrition risk. MST showed worst wt, BMI, TSF, MAC. Higher number of females MNA & MST had the highest specificity (84.6%). sensitivity, indicating at nutritional risk, who limited usefulness in were older, had lower wt, MST had the lowest sensitivity (58.5%). Quality hospitalized elderly. TSF & MAC. score ¼ 20. MST was identified as a SGA was used as a wt change, dietary intake, MST identified 27.4% at benchmark. MST resulted valid screening tool for risk; 72% at no risk. SGA GI symptoms, functional nurses to identify in sensitivity of 94%; identified 20.6% at risk; impairment, muscle malnutrition risk in the specificity of 89%; PPV wasting, fat loss, edema & 79.4% at no risk. elderly subject population. 0.70, indicating high ascites. validity. Quality score ¼ 19. MUST identified as more ht, wt, BMI 63% screened within 6 h of Quality score ¼ 10. effective to screen within admission: accuracy was 12 h of admission; did not questioned; Post training compare to the dietitian e 80% screened within assessment and should be 12 h resulted in greater used as a NST only. accuracy. MUST demonstrated fair In hospital mortality ht, wt, BMI, % wt loss At risk n ¼ 87; Medium ability to predict inincreased with risk n ¼ 25; High risk hospital mortality and los malnutrition risk. n ¼ 62. 3 months ¼ 12%, P ¼ 0.01; BMI <20 kg/m (n ¼ 42) At 6 months ¼ 21%. LOS and included significant increased with wt loss 3e6 months (n ¼ 52) Patients (n ¼ 52) malnutrition risk. Low risk 15d, 95% CI; medium risk at risk or malnourished 24d, 95% CI; High risk e had a lower BMI and 28d 95% CI, P ¼ 0.02. significant weight loss Quality score ¼ 15 3e6 months

5

(continued on next page)

B.L. Cascio, J.V. Logomarsino / Geriatric Nursing xx (2017) 1e8

Martins et al. 200614

Parameters to validate NST

NST: nutrition screening tool, SGA: Subjective Global Assessment, MUST: Malnutrition Universal Screening Tool; MNA: Mini Nutrition Assessment; MNA-SF: Mini Nutrition Assessment Short Form; NRS: Nutrition Risk Screening; GNRI: Geriatric Nutrition Risk Index; Malnutrition Screening Tool CC: calf circumference; MAC: mid arm circumference; MUAC: mid upper arm circumference; alb: albumin; tp: total protein; tlc: total lymphocyte count; pab: prealbumin; rbp: retinol binding protein; hgb: hemoglobin; crp: c reactive protein; LOS: length of stay; wt: weight; ht: height; BMI: body mass index; p: p-value; se: sensitivity; sp: specificity; RCT: randomized controlled trial; PPV: positive predictive value.

GNRI demonstrated greater ability to identify malnutrition risk and related complications than MNA and MNA-SF. In hospital mortality MNA p ¼ 0.003, 3 month: p < 0.001, 6 month: p ¼ 0.002; MNA-SF in hospital mortality p ¼ 0.01; 3 month: p ¼ 0.001; 6 month: p ¼ 0.006. GNRI in hospital mortality: p ¼ 0.038; 3 month: P < 0.001; 6 month: p¼<0.001. Quality score ¼ 14. Low GNRI scores were associated to low MNA, MNA-SF scores including wt, BMI,MAC,CC and alb wt., BMI, MAC,CC; tp, alb, pab. Anthropometric; Biochemical. Prospective Cohort; 131 subjects; Mean age 69.32  8.17 years; October 2011eSeptember 2012. Egypt MNA, MNA-SF, GNRI; <48 h El Gawad et al. 20137

Variables compared Parameters to validate NST Design/Sample size/ Duration/Location Screening assessment/ Screening within 72 h Author

Table 3 (continued )

Malnutrition risk

Outcome

B.L. Cascio, J.V. Logomarsino / Geriatric Nursing xx (2017) 1e8

Statistical significance/ Downs and Black quality score

6

population, relying on recall data from patients is not suggested because of age-associated factors.17 A cross sectional study in Portugal evaluated the effectiveness of the Malnutrition Screening Tool and Nutritional Risk Screening, the nutrition assessment tools Mini Nutrition Assessment and Subjective Global Assessment, and anthropometric parameters compared to Nutritional Risk Screening, and if there was a positive association to increased length of stay.14 Subjective Global Assessment was used as the established control criterion to rate patients’ nutritional status. Patients were classified in one of four categories: well nourished, moderately, undernourished or severely undernourished.14 Anthropometric measurements that were used included: wt, ht, tricep skinfold thickness, and mid-arm circumference. Functional status and length of stay were also evaluated to assess risk of malnutrition. Reliability was established using anthropometric data in addition to Malnutrition Screening Tool, Subjective Global Assessment, and Mini Nutrition Assessment compared to Nutritional Risk Screening. A higher number of females were noted to be at nutritional risk. This observation may have been due to a higher number of female participants (69% women vs 30.9% men).14 Malnutrition Screening Tool correctly identified more patients who were at nutritional risk. The Malnutrition Screening Tool had a similar specificity to Mini Nutrition Assessment (84.6%). However, Malnutrition Screening Tool had the poorest sensitivity performance (58.5%) compared to all the tools examined.14 Nutritional Risk Screening demonstrated to be the best NST in this study to determine risk of malnutrition and increased length of stay. In a randomized control trial conducted in New Zealand, the Malnutrition Screening Tool was compared with Subjective Global Assessment to identify elderly risk of hospital readmission.15 Malnutrition Screening Tool was evaluated for validity for use in the elderly. Malnutrition Screening Tool had been validated for use in inpatients and oncology outpatients in Australia but had yet to be validated with the elderly.19 Patients who met inclusion criteria were screened by two RNs to ensure that sufficient data were present and to minimize interviewer bias. Subjective Global Assessment was shown to have high reliability and validity in use with older adults and was effective at predictive mortality in older hospitalized patients.15 The majority of participants were women (77.1% vs 22.9%). Patients were classified as either malnourished or well-nourished. Malnutrition Screening Tool resulted in a high sensitivity and specificity, indicating that the Malnutrition Screening Tool was a valid NST.15 Subjective Global Assessment identified 20.6% as malnourished compared with 27.6% according to Malnutrition Screening Tool. There was only 1.3% of patients identified as malnourished by Subjective Global Assessment that were not identified in Malnutrition Screening Tool.15 Malnutrition Screening Tool was shown to be a fast and simple screening tool for nurses to determine nutritional risk in the elderly population. Discussion Mini Nutrition Assessment-Short Form showed overestimation of malnutrition risk, suggesting that it was a less effective NST.11,12 Malnutrition Universal Screening Tool, Malnutrition Screening Tool, Nutritional Risk Screening, and Geriatric Nutrition Risk Index showed some NST effectiveness6,11e17 but due to the different variables used to measure risk, determination of accuracy could not be established. All of the studies included large sample sizes, over 100 subjects, which was beneficial. There was limited location bias because the studies were located in various parts of the world: Egypt,7 Indonesia,12 Korea,11 Switzerland,13 New Zealand,15 Portugal14 and the United Kingdom16,17 However, the populations were identified as high-risk based on the subjects’ hospitalization.

IdenƟficaƟon

B.L. Cascio, J.V. Logomarsino / Geriatric Nursing xx (2017) 1e8

Records idenƟfied through database searching (n = 1089)

7

AddiƟonal records idenƟfied through other sources (n =151)

Eligibility

Screening

Records aŌer duplicates removed (n = 1140)

Records screened (n =1140)

Full-text arƟcles assessed for eligibility (n =26)

Did not meet screening criteria Screening greater than 72 hours (n=7) Case studies (n=3) SystemaƟc Reviews (n=2) Review arƟcles (n=4) LeƩer (n=2)

Included

Studies included in qualitaƟve synthesis (n =8)

Records excluded (n =1114)

Fig. 1. Research methodology. From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. Plops Med 6(7): e1000097. http://dx.doi.org/10.1371/journal.pmed1000097

This may have negatively affected the data and accounted for an overestimation of malnutrition based on acute conditions that required hospitalization.18,20 Albumin, a frequent marker that determines protein stores and is widely used to diagnose malnutrition, declines rapidly with infection and illness.20 There were limitations in the NST designed for the elderly. Mini Nutrition Assessment-Short Form is the only NST specifically designed for the elderly; overestimation of malnutrition occurred when Mini Nutrition Assessment-Short Form was used.2,7,11e13,15 Malnutrition Universal Screening Tool11,16,17 and Nutritional Risk Screening11,13 identified more patients at nutritional risk; however, they have not been well researched with the elderly hospitalized population. Another important concern was that healthcare workers conducting screenings were not always fully educated on accurate screening methods.2,8 Nurses are most often the front line staff conducting screenings. Nurses are sometimes required to perform complicated screenings that include giving a subjective opinion on observation of subcutaneous fat loss and muscle wasting. A high patient to nurse ratio and the pressures of completing a large volume of work may prevent nurses from conducting a complicated screening for malnutrition. This indicates the importance of identifying a useful screening tool that is quick and accurate for the identification of malnutrition risk. With a marked growth in the elderly population, staffing for hospitals has not always increased to meet the demand. This has resulted in longer time durations until screening, missed data, and high patient to practitioner ratios. There is no universal screening tool that has been proven effective for the elderly.

Take away messages The five screening tools evaluated in the review do not demonstrate consistent results at identifying malnutrition with complete validity or reliability in the hospitalized elderly. Malnutrition is heightened in the elderly at-risk population which requires early screening and timely intervention to ensure reduced risk of morbidity and mortality. A universal screening system should be implemented along with continual staff training to ensure that health care providers are able to determine at-risk elderly or those already malnourished. Nutritional Risk Screening, Geriatric Nutrition Risk Index, Malnutrition Screening Tool and Malnutrition Universal Screening Tool demonstrated similar effectiveness in identifying patients at risk of malnutrition. The Malnutrition Screening Tool has an advantage of having the quickest administration time. Conclusion This systematic review indicated that the five NSTs that were evaluated showed some effectiveness in identifying malnutrition risk. Based on the lack of a current “gold standard” to screen for malnutrition risk, NSTs have not been shown to prove validity and reliability. The variation in effectiveness between NSTs evaluated in this review suggests that implementation of one specific NST would be difficult. The NSTs evaluated in this review showed similar effectiveness in identifying patients at risk for malnutrition; however each had

8

B.L. Cascio, J.V. Logomarsino / Geriatric Nursing xx (2017) 1e8

limitations. The NSTs evaluated lacked a uniform set of parameters to compare them. Mini Nutrition Assessment-Short Form consistently showed overestimation of malnutrition risk in the elderly.2,6,11e13 Malnutrition Universal Screening Tool was a useful NST in identifying malnutrition risk, predictive validity of inhospital mortality, and hospital length of stay. Nutritional Risk Screening13 was effective at identifying malnutrition in acute illness. However, Nutritional Risk Screening and Malnutrition Universal Screening Tool have limited research in the hospitalized elderly. Malnutrition Screening Tool demonstrated effectiveness compared to Subjective Global Assessment in one study,15 but resulted in poor sensitivity in another study.14 Geriatric Nutrition Risk Index was a better screening tool to identify malnutrition and related complications, but there is limited research on this tool.7 All of the studies reviewed had different parameters, so a true association between them was not identifiable. Future research is suggested with a focus on applying NSTs to the hospitalized elderly using identical parameters and comparing results. Factors such as age, acute condition, and age-related comorbidities should be accounted for to determine the most effective NST. A universal NST can be effective to reduce patient mortality and morbidity, decrease re-admissions, and decrease hospital length of stay.21,22 This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References 1. Omran M, Morley J. Assessment of protein energy malnutrition in older persons, part I: history, examination, body composition and screening tools. Nutrition. 2000;16(1):50e63. http://dx.doi.org/10.1016/s0899-9007(99)00224-5. 2. Bauer J, Kaiser M, Sieber C. Evaluation of nutritional status in older persons: nutritional screening and assessment. Curr Opin Clin Nutr Metab Care. 2010;13(1):8e13. http://dx.doi.org/10.1097/mco.0b013e32833320e3. 3. World Health Organization. What Is Malnutrition?. Available at: http://www. who.int/features/qa/malnutrition/en/. Accessed October 31, 2016. 4. Corkins M, Guenter P, DiMaria-Ghalili R, et al. Malnutrition diagnoses in hospitalized patients: United States, 2010. JPEN J Parenter Enter Nutr. 2013;38(2): 186e195. http://dx.doi.org/10.1177/0148607113512154. 5. Elia M, Bistrian BR. The economics of malnutrition. Nestle Nutr Workshop Ser Clin Perform Programme. 2009;27(2):39e41. http://dx.doi.org/10.1111/j.17534887.1969.tb04950.x. 6. Vanderwee K, Clays E, Bocquaert I, et al. Malnutrition and nutritional care practices in hospital wards for older people. J Adv Nurs. 2010;67(4):736e746. http://dx.doi.org/10.1111/j.1365-2648.2010.05531.x.

7. Abd-El-Gawad W, Abou-Hashem R, El Maraghy M, Amin G. The validity of geriatric nutrition risk index: simple tool for prediction of nutritional-related complication of hospitalized elderly patients. Comparison with mini nutritional assessment. Clin Nutr. 2014;33(6):1108e1116. http://dx.doi.org/ 10.1016/j.clnu.2013.12.005. 8. Mitchell M, Duerksen D, Rahman A. Are housestaff identifying malnourished hospitalized medicine patients? Appl Physiol Nutr Metab. 2014;39(10):1192e 1195. http://dx.doi.org/10.1139/apnm-2014-0133. 9. Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7):e1000097. http://dx.doi.org/10.1371/journal.pmed1000097. 10. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomized and non-randomized studies of health care interventions. J Epidemiol Community Health. 1998;52:377e384. 11. Baek MH, Heo YR. Evaluation of the efficacy of nutritional screening tools to predict malnutrition in the elderly at a geriatric care hospital. Nutr Res Pract. 2015;9(6):637. http://dx.doi.org/10.4162/nrp.2015.9.6.637. 12. Mayasari MS, Lestariana W. Simple Nutritional Screening Tool (SNST) has good validity to identify risk of malnutrition on hospitalized elderly patients. Pak J Nutr. 2014;13(10):573e578. http://dx.doi.org/10.3923/pjn. 13. Drescher T, Singler K, Ulrich A, et al. Comparison of two malnutrition risk screening methods (MNA and NRS 2002) and their association with markers of protein malnutrition in geriatric hospitalized patients. Eur J Clin Nutr. 2010;64(8):887e893. http://dx.doi.org/10.1038/ejcn.2010.64. 14. Martins Catia Patricia Alho Letra, Correia Joana Rita, do Amaral Teresa Freitas. Undernutrition risk screening and length of stay of hospitalized elderly. J Nutr Elder. 2006;25(2):5e21. 15. Wu M-L, Courtney MD, Shortridge-Baggett LM, Finlayson K, Isenring EA. Validity of the malnutrition screening tool for older adults at high risk of hospital readmission. J Gerontol Nurs. 2012;38(6):38e45. 16. Farrer K, Donaldson E, Blackett B, et al. Nutritional screening of elderly patients: a health improvement approach to practice. J Hum Nutr Diet. 2013;27(2):184e191. http://dx.doi.org/10.1111/jhn.12073. 17. Stratton R, King C, Stroud M, Jackson A, Elia M. ‘Malnutrition Universal Screening Tool’ predicts mortality and length of hospital stay in acutely ill elderly. Br J Nutr. 2006;95(02):325. http://dx.doi.org/10.1079/bjn20051622. 18. Stechmiller J. Early Nutritional screening of older adults. J Infus Nurs. 2003;26(3):170e177. http://dx.doi.org/10.1097/00129804-200305000-00009. 19. 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):458e464. http://dx.doi.org/10.1016/s0899-9007(99)00084-2. 20. Sullivan DH, Sun S, Walls RC. Protein-energy undernutrition among elderly hospitalized patients: a prospective study. JAMA. 1999;281(21):2013e2019. http://dx.doi.org/10.1001/jama.281.21.2013. 21. Deutz N, Matheson E, Matarese L, et al. Readmission and mortality in malnourished, older, hospitalized adults treated with a specialized oral nutritional supplement: a randomized clinical trial. Clin Nutr. 2016;35(1):18e26. http://dx.doi.org/10.1016/j.clnu.2015.12.010. 22. Rahman A, Wu T, Bricknell R, Muqtadir Z, Armstrong D. Malnutrition matters in Canadian hospitalized patients: malnutrition risk in hospitalized patients in a tertiary care center using the malnutrition universal screening tool. Nutr Clin Pract. 2015;30(5):709e713. http://dx.doi.org/10.1177/0884533615598954.