Assessing the older person - MNA® Elderly

The Journal of Nutrition, Health & Aging©Volume 7, Number 1, 2003THE JOURNAL OF NUTRITION, HEALTH & AGING©ASSESSING THE OLDER PERSON: IS THE MNA A MORE APPROPRIATENUTRITIONAL ASSESSMENT TOOL THAN THE SGA?L. BARONE, M. MILOSAVLJEVIC, B. GAZIBARICHCorrespondence: Lilliana Barone BSc Master Nut Diet (Sydney University) Clinical Dietitian, Wollongong Hospital Nutrition Department, P.O Box 178, Wollongong, New SouthWales, Australia 2500. e-mail address : baronel@iahs.nsw.gov.auAbstract: OBJECTIVES: A study was undertaken to determine which nutritional assessment tool would bebetter in assessing changes in nutritional status over time in hospitalised older patients. The two tools used werethe Mini Nutritional Assessment (MNA) and the Subjective Global Assessment (SGA). DESIGN: Single blind,prospective study conducted over a 60-day period. SETTING: Five regional hospitals within Southern NewSouth Wales, Australia. SUBJECTS: There were 43 patients at the commencement of the study, then 28 patientsat day 30 and 20 patients at day 60. METHODS: All patients over 65 years of age admitted to the five hospitalsduring May 2001 were eligible. The only exclusions were those patients admitted to palliative care or with severedementia. Two dietitians saw each patient. Each dietitian assessed the patient using his or her assigned nutritionalassessment tool either the SGA or the MNA. All dietitians were randomly assigned the tool at thecommencement of the study. They were familiarised with the use of the tool by participating in a workshop priorto the start of the study. RESULT: The MNA was able to detect greater numbers of malnourished subjects whencompared to the SGA. This finding was consistent across Day 0, 30 and 60 and statistically significant (p

The Journal of Nutrition, Health & Aging©Volume 7, Number 1, 2003ASSESSING THE OLDER PERSONnutritional status of individuals aged 65 years and over?2. Is there a difference in the detection of malnutrition atDay 30 and Day 60 after admission, between the SGA andMNA?MethodsSubjectsThe study included patients 65 years of age and overadmitted to five hospitals within the IAHS. All new admissionsto these sites over a one-month period (between May and June2001) were eligible for the study; with the exception ofpalliative care patients and patients with severe dementia.DesignThis was a prospective single blind study conducted over a60-day period. A paired sample was used. Two observations(the SGA and MNA) were done on each participant at baseline(Day 0), Day 30 and Day 60 after admission. Those patientsthat were discharged during this period were followed up athome or place of residence. Results and other relevant datasuch as social status, and admitting diagnosis (if available)were recorded on a simple patient data form by each observerfor every subject.Tools for AssessmentThe SGA is a subjective global assessment tool designed forhospitalised patients (11,12) and the MNA is specificallydeveloped for the assessment of the nutritional status of theolder person (16,17).Modification of the MNA toolThe MNA was developed in France with collaboration fromthe Clinical Nutrition Program, University of New MexicoUSA and the Nestle Research Centre, Lausanne Switzerland(18). After consultation with the author (19), the MNA tool wasmodified as follows to make it more culturally specific to theolder Australian population.i) Question F Body Mass Index (BMI) (weight inkg)/(height in m) 2Score BMI Original MNA BMI Modified MNA0 Less than 19 Less than 221 19 to less than 21 22 to less than 24.52 21 to less than 23 24.5 to less than 273 23 or greater 27 or greaterThe French anthropometric standards are lower than thoseused for the USA and Australia, therefore the BMI range wasmodified to reflect the ranges recommended by the DietaryGuidelines for Older Australians. For healthy older adults mostassessment standards suggest a range of 22 - 27 (20), thereforeQuestion F in the screening component was adjusted asillustrated above.ii) Question K Selected consumption markers forprotein intakeThe original MNA had two or more servings of legumes oreggs per week. After consultation with the author, this wasmodified to one - two servings per week to better match theAustralian dietary practices. Legumes are traditionallyconsumed in greater quantities in Mediterranean diets. Onaverage 18g legumes are consumed per day in Mediterraneandiets compared with 6g per day in the Australia (21,22).iii) Question Q Mid-arm circumference (MAC) in cmScore MAC Original MNA MAC Modified MNA0.0 MAC less than 21 MAC less than 26.6(male)25.1(female)0.5 MAC 21 to 22 MAC 26.6 to 30.4 (male)25.1 to 29.7 (female)1.0 MAC 22 or greater MAC 30.4 or greater(male)39.7 or greater (female)Adjustments to MAC measurements were made based on theAmerican percentile norms for measurements of MAC in olderwhite subjects (23), which also included gender differences.ProcedureFor the purpose of the study, participating hospitals weregrouped into three sites. A total of six dietitian’s (two fromeach site) acted as investigators. The investigators worked inpairs, each being randomly assigned a nutrition assessment tool– either the SGA or MNA.All investigators had been trained in the use of the SGA andwere currently using this tool in their normal clinical practice.To minimise user bias and familiarise themselves with theMNA tool, all dietitians involved in the study used the MNAover a trial period of one week prior to the commencement ofthe study. To maximise inter-rater reliability a workshop wasconducted prior to this trial where the dietitians were trained inthe use of the MNA.ImplementationThe SGA and MNA were each conducted on all new (andappropriate) admissions, over a 30-day period. The firstassessment on admission was Day 0 (baseline). The sameinvestigators used the same assessment tool to reassess eachsubject at Day 30 and again at Day 60. “Patient data forms”were completed for each subject at Day 0, 30 and 60, andforwarded on to another dietitian, independent of the study forcollation. If a subject was identified as nutritionallycompromised the site dietitian provided nutrition carethroughout their admission as per standard dietetic practicewithin the Nutrition Department. Follow-up of the subjects atDay 30 and 60 were arranged closer to the time due. Wheresubjects were willing, an outpatient clinic appointment wasmade, however, the majority of follow-ups were home visits.14

The Journal of Nutrition, Health & Aging©Volume 7, Number 1, 2003THE JOURNAL OF NUTRITION, HEALTH & AGING©ResultsOverall 43 subjects were assessed at baseline using eachtool. Table 1 describes the subjects’ characteristics at the threetime points. There was no statistical difference in the variablesat each of the three time points measured.Figure 2Nutritional status as determined by the SGAand MNA at Day 30Table 1Demographic characteristics of subjectsVariables Measured Baseline Day 30 Day 60Number of Subjects 43 28 20Age (years) mean 78.6 78.2 80(range) (68-94) (68-92) (70-92)Sex % female 60 50 55% male 40 50 45BMI (kg/m2) mean 24.5 23.8 24.2Female 24.8 23.9 24.4Male 24.0 23.7 24.0Social Status % alone 35 50 50% partner/carer 60 50 50% hostel 5 0 0% nursing home 0 0 0Figure 3Nutritional status as determined by the SGAand MNA at Day 60Reasons for the dropout included subjects withdrawal fromthe study, subjects uncontactable for follow up, subjects onholiday over the follow up period, difficulty with time/resourceconstraints for dietitians to follow up some subjects whorequired home visits.Graphs (Figure 1,2 and 3) illustrate the nutritional status asdetected by the SGA and MNA over the 60-day time period.There was not a significant difference between the numbersidentified as undernourished (i.e. at risk and malnourished)with each tool. The significance however, lies in the degree ofmalnutrition detected, with the MNA identifying a greaterproportion of malnourished subjects consistently across thetime intervals, compared to the SGA.Figure 1Nutritional status as determined by the SGAand MNA at Day 0DiscussionThe literature cites rates of malnutrition amongsthospitalized older persons as ranging from 20 to 60% (16,24).Similar to the rates identified in this study. The rates ofmalnutrition and those at risk of malnutrition identified in thisstudy using the modified MNA reflect similar rates to thosefound in other MNA studies (25).The major finding from this study is that the modified MNAdetected greater numbers of malnourished subjects whencompared to the SGA. This finding was consistent across thethree time intervals measured and was statistically significant(see table 2).15

The Journal of Nutrition, Health & Aging©Volume 7, Number 1, 2003ASSESSING THE OLDER PERSONTable 2Percentage of Subjects Detected as Malnourished using theSGA and MNA assessment toolsAssessment Tool/Time period SGA MNA *P value(category c) (less than 17)Day 0 ( n=43) 5 21 0.005Day 30 ( n=28) 4 18 0.005Day 60 ( n=20) 0 20 0.05* Analysed using the Wilcoxon-Sign Ranked Test.This finding supports the claim by other authors that theSGA cannot be used to monitor changes in nutritional statusbecause of its subjectivity and non-quantitative data analysis(17), whereas the quantitative nature of the MNA allows foreasier monitoring of nutritional changes over time. Otherstudies using the SGA for nutritional assessment of the olderperson have found discrepancies in the detection ofmalnutrition (14). This may be explained in part by the lowerinter-rater agreement found with the SGA when used with olderpatients (13,14). The MNA however, has been found to havegood inter-rater reproducibility (15).One of the advantages of the MNA for dietitians is that itdoes not rely on additional measurements that may be difficultfor dietitians to access- such as blood tests. It is a tool that issensitive enough to detect the small changes in nutritionalstatus that may occur over time with nutritional support.Further work is needed to determine the extent to which theMNA can be used as a detector of nutritional change over time.Questions such as, “At what time interval should a patient havea repeat nutritional assessment performed to most accuratelydetect altered nutritional status?” need to be investigated.Vellas (17) suggests that monthly weight measurements and arepeat MNA at 3 months is an important follow up fornutritionally compromised older patients. A nutritionintervention study conducted in Nursing Homes in France (25)demonstrated changes in MNA score at day 60 following oralsupplementation.It is also interesting to note that the average BMI of thisstudy population was 24.5 (within the normal standard range of22-27). The MNA (at a score between 17 and 23.5) hasdemonstrated an ability to assess declining i.e. “at risk”nutritional status before severe changes in weight (or albuminlevels) occurs (15).Future Directions for Clinical PracticeThe ability of the modified MNA to more accurately identifymalnourished older patients compared to the SGA asdemonstrated by these study findings has importantimplications for dietetic practice and care of this group.Nutrition intervention is said to be easier and more effectivewhen implemented in those identified as at risk of malnutritionthan in those who are severely malnourished (26). Furtherinvestigation is warranted to determine if changes in currentdietetic practices, such as more aggressive intervention innutritional support, are necessary.The screening component (MNA-SF) (27) could be easilyadministered by nursing/medical staff on admission to hospital,with those identified as at risk of malnutrition (MNA screeningscore 11 points or below) being referred to the dietitian forfurther assessment and appropriate nutrition intervention. Thiswould increase staff awareness of the prevalence ofmalnutrition amongst older patients and would also allowdietitians to spend more of their clinical time in nutritionassessment, intervention and follow up of these patients.ConclusionIn this study, the modified MNA has proved to be a moreappropriate nutrition assessment tool than the SGA inhospitalised older patients. It is better able to identifymalnourished patients. This could potentially impact on dieteticpractice, as more aggressive nutritive therapy may be necessaryto improve nutritional outcomes within this patient group.Acknowledgments: We would like to thank the following dietitians for their invaluableassistance in the reviewing of the manuscript and their involvement in the data collection.Without their assistance and cooperation this work could not have been done. Kate Burge,Jodie Crouch, Jennifer Haughton, Jacqueline Hampton, Kelly Lambert, Craig Patch andJane Wishart.References1. Edington, J., Kon P., Martyn,C.N.,Prevalence of malnutrition in patients in generalpractice, Clin. Nutr., 1996;15:60-63.2. SENECA, Investigators. Longitudinal changes in iron and protein status of olderperson Europeans, Eur. J. Clin. Nutr. 1996;50(suppl:2):S16-S24.3. Cederholm, T., et al., Nutritional status and performance capacity in internal medicalpatients. Clin. Nutr., 1993,12:8-14.4. Morley, J.E., Silver, A.J., Nutritional issues in nursing home care, Ann. Intern.Med.,1995;123:850-859.5. Mowe,M., Bohmer, T., Kindt, E., Reduced nutritional status in an older personpopulation (>70 y) is probable before disease and possibly contributes to thedevelopment of disease,Am. J. Clin. Nutr., 1993;58:317-324.6. Kaye,G,L.,In: Outcomes management. Linking research to practice. Joint publicationof ADA and Ross Products division of Abbott Laboratories 1996.7. Reilly, J.J., et al., Economic impact of malnutrition: a model system for hospitalisedpatients, J. Parenter. Enteral. Nutr.,1988;12:371-3768. Beck, A.M., Ovesen,L.,Osler,M., The ‘Mini Nutritional Assessment’ (MNA) and the‘Determine Your Nutritional Health’ Checklist (NSI Checklist) as predictors ofmorbidity and mortality in an older person Danish population, British Journal ofNutrition.., 1999;81:31-36.9. Beck, E.,etal.,Implementation of malnutrition screening and assessment by dietitians:malnutrition exists in acute and rehabilitation settings, Australian Journal Nutritionand Dietetics., 2001;58(2):92-97.10. Ferguson, M., Bauer, J., Banks, M., Capra, S.,Malnutrition screening and assessmentresource manual. Brisbane: FBBC Nutrition Research Group; 1996.11. Detsky,A.S., et al., Evaluating the accuracy of nutritional assessment techniquesapplied to hospitalized patients: methodology and comparisons, Journal of Parenteraland Enteral Nutrition .,1984;8(2):153-9.12. Detsky, A.S., et al., What is Subjective Global Assessment of Nutritional Status?Journal of Parenteral and Enteral Nutrition.,1987;11(1):8-13.13. Deurksen, D.R., et al., The Validity and reproducibility of clinical assessment ofnutritional status in the older person, Nutrition.,2000;16:740-744.14. Ek, A., et al., Interrater variability in subjective global assessment of older personpatients, Scand. J. Caring. Sci., 1996;10:163-168.15. Vellas, B., et al., The Mini Nutritional Assessment (MNA) and its use in grading thenutritional status of older person patients.,Nutrition, 1999;15(2):116-122.16

The Journal of Nutrition, Health & Aging©Volume 7, Number 1, 2003THE JOURNAL OF NUTRITION, HEALTH & AGING©16. Guigoz, Y., Vellas, B., Garry, P., Assessing the nutritional status of the older person:The Mini Nutritional Assessment as part of the geriatric evaluation, NutritionReviews., 1996;54(1):S59-S65.17. Vellas, B., et al., Nutrition Assessment in the older person, Curr. Opin. Clin. Nutr.Metab. Care., 2001;4:5-8.18. Guigoz, Y.,and Vellas, B., The Mini Nutritional Assessment (MNA) for grading thenutritional state of older person patients: Presentation of the MNA, history andvalidation. In Mini Nutritional Assessment (MNA): Research and Practice in theolder person. Nestle Nutrition Workshop Series Clinical & Performance ProgrammeNo. 1 1998. Nestle Nutrition Services:1-2.19. Guigoz., Y., Personal communication. 6 February 2001.20. Bartlett, S., Marian, M., Taren, D., and Muramoto, M., Geriatric Nutrition Handbook1998. Chapman & Hall Nutrition Handbooks 5. International Thomson Publishing.21. Renaud, S.,In. Le Regime Sante 1995. Editions Odile Jacob. Pg 1922. Apparent Consumption of Food stuffs in Australia, 1996-97. (Australian Bureau ofStatistics Catalogue No 4306.0).23. Mitchell, C., and Chernoff, R., Nutritional assessment of the older person in GeriatricNutrition: The Health Professionals Handbook 1991. ASPEN publishers Maryland:chpt 14.24. Saletti, A., etal., Nutritional status according to Mini Nutritional Assessment in aninstitutionalized older person population in Sweden, Gerontology., 2000;46:139-145.25. Lauque, S., et al..,Protein-energy oral supplementation in malnourished nursinghomeresidents. A controlled trial, Age and Ageing .,2000;29:51-5626. Vellas, B., et al.,Relationships between nutritional markers and the Mini NutritionalAssessment in 155 older persons, Journal American Geriatric Society.,2000;48:1300-1309.27. Rubenstein, L.Z., et al., Screening for undernutrition in geriatric practice.Developing the short-form mini-nutritional assessment (MNA-SF),The Journals ofGerontology., 2001;56A(6):M366-72.17