DOI: 10.1002/ncp.

11120

CLINICAL RESEARCH

Global Leadership Initiative on Malnutrition criteria

in older adults who are institutionalized: Agreement
with the Subjective Global Assessment and its impact
on 5‐year mortality

Guilherme F. Graciano RD, MSc1 | Isabella R. Souza RD, MSc2 |

Maria Isabel T. D. Correia MD, PhD3 | Lucilene R. Anastácio RD, PhD2 |
Bárbara C. Santos RD, MSc2

1
Sciences Applied to Adult Health
Graduate Program, Faculty of Medicine, Abstract
Universidade Federal de Minas Gerais, Background: This study aimed to assess the prevalence of malnutrition
Belo Horizonte, Brazil
2
according to Subjective Global Assessment (SGA), Mini Nutritional
Food Science Graduate Program, Faculty
of Pharmacy, Universidade Federal de Assessment–Full Form (MNA‐FF), and different combinations of the Global
Minas Gerais, Belo Horizonte, Brazil Leadership Initiative on Malnutrition (GLIM) criteria in older adults who are
3
Surgery Graduate Program, Faculty of institutionalized, and the impact of malnutrition on 5‐year mortality.
Medicine, Universidade Federal de Minas
Methods: Nutrition status was assessed by the SGA, MNA‐FF, and 15 GLIM
Gerais, Belo Horizonte, Brazil
criteria combinations. The Katz scale was used to assess the level of
Correspondence dependence. The SGA was considered the reference method, and the
Bárbara C. Santos, RD, MSc, Food Science
Graduate Program, Faculty of Pharmacy,
agreement (Kappa test), sensitivity, and specificity values were calculated for
Universidade Federal de Minas Gerais, each GLIM criteria combination. The variables associated with 5‐year
Belo Horizonte, Brazil. mortality were assessed using multivariate logistic regression models.
Email: [email protected]
Results: One hundred eleven participants (mean age: 81y; interquartile range:
76.0–87.0; 90.9% women) were included; the prevalence of malnutrition
according to the SGA and MNA‐FF were 49.5% (n = 55) and 8.1% (n = 9),
respectively. The prevalence of malnutrition varied from 1.8% to 36.0%
considering GLIM combinations. Eight GLIM criteria combinations had a fair
agreement with SGA (κ: 0.21–0.40), and two had sensitivity >80%. Regarding
mortality, 43 participants (38.7%) died within 5 years. Malnutrition according
to the SGA (odds ratio [OR]: 2.82; 95% confidence interval [CI]: 1.06–7.46) and
the Katz scale score (OR: 4.64; 95% CI:1.84–11.70) were independent
predictors of mortality.
Conclusion: The prevalence of malnutrition varied according to the
assessment tools. Malnutrition diagnosed by the SGA, but not by the GLIM
criteria or MNA‐FF, was associated with 5‐year mortality in older adults who
were institutionalized.

© 2024 American Society for Parenteral and Enteral Nutrition.

Nutr. Clin. Pract. 2024;1–8. wileyonlinelibrary.com/journal/ncp | 1

 19412452, 0, Downloaded from https://aspenjournals.onlinelibrary.wiley.com/doi/10.1002/ncp.11120 by Instituto Nacional De, Wiley Online Library on [06/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 | GRACIANO ET AL.

KEYWORDS
GLIM criteria, homes for the aged, malnutrition

I N T R O D U C TI O N Therefore, this study aimed to assess the prevalence of

malnutrition according to different methods (GLIM,
The aging process is associated with several physical, MNA, and SGA) and the agreement between the tools
psychological, nutrition, and social changes.1 Older as well as the impact of malnutrition on the 5‐year
adults often exhibit reduced dietary intake because of mortality in older adults who are institutionalized.
many factors that are associated with the common
presence of acute and/or chronic diseases that contribute
to the impairment of nutrition status. Furthermore, ME T H OD S
malnutrition is associated with poor prognosis in older
persons in all settings (those in hospitals, those in long‐ Study design, setting, and participants
term care facilities, and those who are community‐
dwelling).1 This was a secondary analysis of a cross‐sectional study
In general, >40% of older adults in any healthcare including older adults living in long‐term care facilities in
setting are at risk for malnutrition according to the the Belo Horizonte (Minas Gerais, Brazil) who were initially
Mini Nutritional Assessment (MNA)–Short Form, and assessed regarding their nutrition, sociodemographic, and
the highest rates of malnutrition are observed in older clinical statuses and had mortality rates assessed 5 years
adults who are institutionalized (approximately 30%).2 after the initial assessment. The data used in this study
Between 30% and 80% of older adults who are were collected by a trained dietitian, and all participants
institutionalized have some nutrition deficiency,3 and were assessed during the same period.
the prevalence of nutrition risk can be as high as 95%.4 Both female and male residents were screened for
Poor clinical outcomes, with an increased incidence of inclusion according to the following inclusion criteria:
infectious complications and hospital length of stay, are living in one of the four long‐term care facilities, being
consequences of malnutrition.1 ≥60 years of age, and agreeing to participate in the study
In this context, an adequate diagnosis of malnutrition by signing the consent form. The exclusion criteria were
is essential to enable early nutrition interventions. There as follows: receiving enteral feeding and having a lack of
are several nutrition assessment tools; in particular, for interest in participation or consent from the legal
older adults, the MNA5 encompasses the assessment of guardians. All participants who were institutionalized
food intake, weight loss, mobility, the presence of acute during the study period and met the criteria for inclusion
disease, and neuropsychological alterations to diagnose were evaluated.
malnutrition. On the other hand, the Subjective Global The study was approved by the ethics committee
Assessment (SGA),6 a clinical method validated world- (protocol number 51311715.3.0000.5149). The primary
wide, can also be used in older persons. More recently, data collection of the cross‐sectional study occurred
the Global Leadership Initiative on Malnutrition (GLIM) between February 2016 and August 2016. The assess-
criteria were published, aiming to operationalize and ment of mortality data occurred 5 years later, between
standardize malnutrition diagnoses around the world. January 2021 and April 2021.
According to the GLIM criteria, the combination of
phenotypic criteria, such as nonvolitional weight loss,
low body mass index (BMI), and low muscle mass Variables
(identified by any validated method), with etiologic
criteria (reduced food intake or nutrient assimilation Only one trained dietitian collected the variables. Mortality
and disease burden or inflammatory condition) provides data were obtained from the records of the institutions,
the diagnosis of malnutrition, and at least one pheno- and death certificates were obtained by telephone contact
typic and one etiologic criterion are required.7 with the institutions’ managers. Sociodemographic and
Considering the high risk for malnutrition in older clinical data were collected, and the nutrition status of all
adults who are institutionalized, the clinical impact of participants was assessed by different methods, such as the
malnutrition in this population, and the proposed GLIM MNA–Full Form (MNA‐FF), the SGA, anthropometry, and
criteria as a tool to standardize the malnutrition diagnosis, food intake. Furthermore, the Portuguese version of the
the comparison of the GLIM criteria with validated tools Katz scale8 was used to verify independence in basic
and the assessment of their predictive value is necessary. activities of daily living. The Katz score ranges from 0 to 6,
 19412452, 0, Downloaded from https://aspenjournals.onlinelibrary.wiley.com/doi/10.1002/ncp.11120 by Instituto Nacional De, Wiley Online Library on [06/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
NUTRITION IN CLINICAL PRACTICE | 3

with higher values indicating higher levels of dependence. weight minus the preconsumption weight of the meal.
The sociodemographic and clinical variables were age, sex, The dietary data were assessed regarding the total daily
ability to walk or dependence on a wheelchair, presence of energy intake calculated using the Brazilian Food
chronic diseases (chronic obstructive pulmonary disease, Composition Table13 and Microsoft Excel software. The
Alzheimer disease, Parkinson disease, diabetes, hyper- estimated energy requirement was calculated considering
tension, or cancer), and smoking history. current weight, sex and the activity factor,14 which was
According to their MNA‐FF scores, patients were equal to 1.0 for all participants because all had the
classified as follows: nourished (≥24), at risk for minimum physical activity level.
malnutrition (between 17 and 23.5), and malnourished
(<17).5 According to the SGA, SGA A patients were
classified as nourished, SGA B patients were classified as GLIM criteria and mortality data
potentially or moderately malnourished, and SGA C
patients were classified as severely malnourished.6 The GLIM criteria encompass at least one phenotypic
Patients classified as SGA B or C were grouped as and one etiologic criteria.7 In this study, the phenotypic
malnourished for statistical purposes. For the anthropo- criteria were low BMI (age < 70 years: <20 kg/m²;
metric assessment, the following measures were taken: age >70 years: <22 kg/m²) and low muscle mass esti-
weight, height, BMI, midarm circumference (MAC), mated by the low MAMA (≤15th percentile)15 and low
triceps skinfold thickness (TSF), and calf circumference CC (men: ≤34 cm; women: ≤33 cm).12 The etiologic
(CC). The values for MAC and TSF were used to criteria were disease burden (classified by the Charlson
calculate the midarm muscle area (MAMA).9 The weight Comorbidity Index as ≥116) and low food intake (energy
and height were estimated10 when it was not possible to intake ≤50% of the energy requirement).7 Considering
assess. In participants with amputated limbs, a correction these criteria, 15 different GLIM criteria combinations
was performed.11 For circumferences and TSF, the mean were obtained and used in the analysis. The mortality
value of three measurements was considered. data were obtained by phone contact with the institu-
The nutrition diagnosis of the patients was performed tions, and the deaths up to 5 years after the assessment
using the MNA and the SGA. Body weight was measured were considered in the analysis.
using a digital scale with a maximum capacity of 150 kg.
When it was not possible to weigh the participant, weight
was estimated using CC, arm circumference, and Statistical methods
subscapular skinfold thickness values. In the case of
amputees, the weight corresponding to the amputated The results were expressed as the median and inter-
limb was subtracted from the ideal weight. Arm quartile range (IQR). Kappa values were used to assess
circumference was measured with the participants’ the agreement between the diagnosis of malnutrition
relaxed arms hanging by their sides, at the midpoint according to the different combinations of the GLIM
between the acromion and olecranon. criteria and the SGA, which was considered the
CC was measured in a seated or supine position for semi–reference standard in this study. The agreement
bedridden participants, with the leg at a 90° angle, at its was classified as very good (κ: >0.80), good (κ: 0.61–0.80),
most prominent point. CC was considered low if it was moderate (κ: 0.41–0.60), fair (κ: 0.21–0.40), or poor
≤34 cm for men and ≤33 cm for women, according to (κ: <0.20).17 Furthermore, the sensitivity, specificity,
Brazilian cutoff data.12 Circumferences were measured and positive and negative predictive values were calcu-
using a nonstretchable tape performed in triplicate, and lated by comparing each GLIM criteria combination with
the mean of the values was used. All measurements were the SGA. Univariate and multivariate logistic regression
taken on the right side of the body.13 models were used to verify the variables associated with
The dishes, cups, bowls, and mugs were identified higher odds for mortality in 5 years. The variables with a
and weighed, and the liquid and solid meals were P value <0.2 in the univariate analysis were included in
weighed individually using a culinary scale with a the multivariate analysis. Models adjusted by sex and age
maximum capacity of 5 kg (Black and Decker, Model were tested for each tool (GLIM criteria combinations,
BC500). The amount of food ingested was individually SGA, and MNA‐FF). The model adjustment was verified
calculated in advance for each participant by a dietitian. by the Hosmer‒Lemeshow test as >0.05. The software
The older adults consumed only the meals provided by Statistical Package for Social Sciences (SPSS Inc) version
the institution. The food intake was assessed by 21.0 was used for the analysis. A P value <0.05 was
direct weighing on 3 nonconsecutive days, including considered significant. Because this is a study with
1 day on the weekend, considering the postconsumption secondary data analysis of a prospective cohort, the
 19412452, 0, Downloaded from https://aspenjournals.onlinelibrary.wiley.com/doi/10.1002/ncp.11120 by Instituto Nacional De, Wiley Online Library on [06/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 | GRACIANO ET AL.

power of the study was calculated. The sample had a T A B L E 1 Sociodemographic characteristics of the other adults
power >80% for demonstrating differences in the in long‐term care.
prevalence of malnutrition according to the SGA Variable n %
(49.5%) and all GLIM criteria combinations. In relation Sex
to mortality, the combination BMI plus disease burden
Female 100 90.1
had a power of 68.6% for demonstrating differences
in mortality. Male 11 9.9
Age
60–69 years old 18 16.2
RESU LTS
70–79 years old 30 27.0

One hundred and eleven older adults were included in ≥80 years old 63 56.8
this study, and female participants corresponded to 90.1% Ethnicity
of the sample because three of the four institutions were
White 75 68.5
exclusively for women (Table 1). The average age was
80.4 ± 9.4 years, with men averaging 72.9 ± 8.5 years and Black 27 25.2
women averaging 81.2 ± 9.1 years (P = 0.005). Most of Mixed race 6 5.4
them were very elderly, with 60.0% being >80 years old, Indigenous 1 0.9
whereas, among men, 54.5% were between 60 and 69
Marital status
years old (P = 0.003).
None of the GLIM criteria combinations had a very Single 69 69.0
good agreement with the SGA. The following combinations Widower 20 20.0
had fair agreement: GLIM BMI plus disease burden, GLIM Other 11 11.0
CC plus disease burden, GLIM CC and/or BMI plus disease
Children
burden, GLIM MAMA and/or BMI plus disease burden,
GLIM BMI and/or CC plus low food intake and/or disease Yes 45 44.1
burden, and GLIM BMI and/or MAMA plus low food No 57 55.9
intake and/or disease burden. All of the remaining Education
combinations had poor agreement. In general, the combi-
Illiterate 36 34.6
nations including low food intake as the etiologic criteria
had a lower prevalence of malnutrition and worse Elementary school incomplete 43 41.3
agreement with the SGA, compared with that of combina- Elementary school complete 12 11.5
tions including disease burden (Table 2). Middle school incomplete 3 2.7
Regarding mortality, 43 participants (38.7%) died
Middle school complete 6 5.3
within 5 years. The variables tested in the univariate
analysis are depicted in the supplementary file. The High school incomplete 1 0.9
models obtained can be found in Table 3. Neither High school complete 3 2.7
the GLIM criteria combination nor the MNA‐FF were Years of institutionalization
able to predict mortality. In the multivariate analysis,
0–5 73 65.8
only malnutrition according to the SGA (odds ratio [OR]:
2.82; 95% confidence interval: 1.06–7.46—model 5) and 0–10 17 15.3
Katz score ≥ 5 (OR between 4.64 and 6.23 in the different 11–15 10 9.0
models depicted in Table 3) were independently associ-
>16 11 9.9
ated with a higher odds of 5‐year mortality in 73.9% of
this sample of older adults who are institutionalized
(model 5).
criteria. When using the GLIM criteria, different
combinations of phenotypic and etiologic criteria yielded
DISCUSSION diverse results for malnutrition prevalence. The agree-
ment between the GLIM criteria and the SGA was poor,
The prevalence of malnutrition in older adults who are and only two combinations had adequate sensitivity.
institutionalized varies widely according to different None of the 15 different GLIM criteria combinations was
used methods, including the SGA, MNA‐FF, and GLIM an independent predictor of 5‐year mortality. On the
 19412452, 0, Downloaded from https://aspenjournals.onlinelibrary.wiley.com/doi/10.1002/ncp.11120 by Instituto Nacional De, Wiley Online Library on [06/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
NUTRITION IN CLINICAL PRACTICE | 5

T A B L E 2 Prevalence of malnutrition according to the different GLIM criteria combinations and kappa values compared with the
Subjective Global Assessment in older adults who are institutionalized.
GLIM criteria n (%) Kappa (P value) SE SP PPV NPV
Body mass index + low food intake 2 (1.8) 0.037 (>0.05) 3.6 51.4 100.0 100.0
Body mass index + disease burden 17 (15.3) 0.311 (<0.001) 30.9 100.0 100.00 59.6
Calf circumference + low food intake 6 (5.4) 0.037 (>0.05) 7.3 96.4 66.7 51.4
Calf circumference + disease burden 36 (32.4) 0.295 (<0.001) 47.3 82.1 72.2 61.3
Midarm muscle area + low food intake 2 (1.8) 0.037 (>0.05) 3.6 100.0 100.0 51.4
Midarm muscle area + disease burden 12 (10.8) 0.183 (0.002) 20.0 98.2 91.7 55.6
Calf circumference and/or body mass index + low food intake 6 (5.4) 0.037 (>0.05) 66.7 51.4 7.3 96.4
Calf circumference and/or body mass index + disease burden 36 (32.4) 0.295 (<0.001) 72.2 61.3 47.3 82.1
Midarm muscle area and/or body mass index + low food intake 3 (2.7) 0.055 (>0.05) 100.0 51.9 5.5 100.0
Midarm muscle area and/or body mass index + disease burden 20 (18.0) 0.329 (<0.001) 34.5 98.2 95.0 60.4
Midarm muscle area + low food intake and/or disease burden 13 (11.7) 0.202 (<0.001) 21.8 98.2 92.3 56.1
Body mass index + low food intake and/or disease burden 18 (16.2) 0.018 (>0.05) 92.3 56.1 21.8 98.2
Calf circumference + low food intake and/or disease burden 40 (36.0) 0.332 (<0.001) 72.5 63.4 80.4 52.7
Body mass index and/or calf circumference + low food intake 40 (36.0) 0.332 (<0.001) 52.7 80.4 72.5 63.4
and/or disease burden
Body mass index and/or midarm muscle area + low food intake 22 (19.8) 0.366 (<0.001) 60.0 57.7 55.9 53.6
and/or disease burden
Abbreviations: GLIM, Global Leadership Initiative on Malnutrition; NPV, negative predictive value; PPV, positive predictive value; SE, sensitivity;
SP, specificity.

other hand, malnutrition according to the SGA was an important to highlight that, in the current study,
independent predictor of mortality as well as lower food intake was assessed by direct weighing, which is
independence, according to the Katz scale. considered a reference method.23 Therefore, the cutoff of
Sanz‐Paris et al18 assessed 12 different GLIM criteria 50%, recommended by GLIM comitee,7 could be very
combinations in 485 older adults who were institutional- rigid for determining low food intake.
ized and observed differences in the prevalence of Only a few combinations of GLIM criteria showed
malnutrition according to the specific combinations of sensitivity or specificity values >80%, and none showed
phenotypic and etiologic criteria. The authors reported a adequate values for both. Only two combinations
prevalence of malnutrition from 5.7% to 19.2%, with an reached sensitivity values >80%, and eight combinations
average prevalence of 13.5% and 10.5% for combinations had a specificity >80%. Studies with hospitalized patients
with low food intake or disease burden (inflammation have shown divergent values of sensitivity and specificity
associated with acute disease) as the etiologic criterion, when comparing SGA and GLIM criteria. Fernandez
respectively. A decrease in food intake for at least 2 et al24 evaluated 165 participants >65 years of age
weeks was reported by the patient or their caregivers and admitted to the emergency ward of a Brazilian hospital.
later confirmed by institutional caretakers.18 This varia- The GLIM criteria (at least one phenotypic and one
tion was also observed in studies with other populations etiologic criteria) showed 76% sensitivity and 75.1%
when the different GLIM criteria combinations were specificity, according to the MNA‐FF, and the usual food
tested.19,20 In our study, a small number of individuals intake of the participants was assessed by applying a
with low food intake (7.2%) were observed, probably dietary history from the last month in which the data
because of the low cutoff (energy intake <50% of the were collected. Allard et al20 evaluated 784 patients from
requirement) and the nutrition assistance offered by the 18 Canadian hospitals and found a sensitivity of 61.3%
institutions included in this research. In other studies and specificity of 89.8%, considering all combinations of
with older adults, higher proportions of low food intake GLIM criteria and the SGA. In this same study, low BMI
(13.5%21 and 52.5%22) were reported. However, it is also plus low intake had the highest specificity (98.8%) but a
 19412452, 0, Downloaded from https://aspenjournals.onlinelibrary.wiley.com/doi/10.1002/ncp.11120 by Instituto Nacional De, Wiley Online Library on [06/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 | GRACIANO ET AL.

T A B L E 3 Logistic Regression Models obtained for predict cancer22 who were moderately or severely malnourished
mortality in older adults who are institutionalized. had significantly lower survival rates. In the study by
Variable OR 95% CI P value Yeung et al26 (n = 1576), weight loss and the presence of
Model 1 (HL: 0.398), 72.1%
disease were risk factors for mortality among older adults
who are institutionalized. In this context, it is important
GLIM criteria: body mass 1.86 0.54–6.34 0.322
to highlight that the present study was conducted in a
index + disease burden
convenience sample, and the number of participants may
Katz score ≥5 5.76 2.35–14.12 <0.001 be insufficient to adequately demonstrate the results,
Model 2 (HL: 0.395), 73.0% especially regarding the power of combinations to predict
GLIM criteria: midarm muscle 1.94 0.41–9.05 0.399 mortality in the predictive analysis. In this regard, this is
area + disease burden one limitation of our study because mortality data were
collected by phone contact with professionals from long‐
Katz score ≥5 6.23 2.58–15.13 <0.001
term care institutions because of the COVID‐19 pan-
Model 3 (HL: 0.580), 72.1% demic, and the exact date of death was not retrieved.
GLIM criteria: midarm muscle 1.88 0.58–6.04 0.288 Therefore, we could only perform logistic regression for
area and/or body mass mortality instead of a Cox regression. Additionally,
index + disease burden objective data regarding weight loss were not available;
Katz score ≥5 5.87 2.41–14.30 <0.001 therefore, this important phenotypic criterion could not
Model 4 (HL: 0.525), 72.1% be assessed. Furthermore, low muscle mass was esti-
mated by anthropometric parameters. However, among
GLIM criteria: body mass index 1.87 0.61–5.68 0.268
the strengths of this study, it can be cited the prevalence
and/or midarm muscle
area + low food intake of malnutrition according to GLIM criteria considering
and/or disease burden 15 combinations of phenotypic and etiologic criteria, and
food intake assessed by direct weighing.
Katz score ≥5 5.86 2.41–14.26 <0.001
Model 5 (HL: 0.367), 73.9%
SGA B/C 2.82 1.06–7.46 0.036 CONCLUSION
Katz score ≥5 4.64 1.84–11.70 0.001
Important variations in the prevalence of malnutrition
Model 6 (HL: 0.866), 73.9%
were observed when different tools and GLIM criteria
MNA‐FF <24 3.02 0.33–27.28 0.325 combinations were tested. Most of these combinations
Katz score ≥5 5.95 2.45–14.44 <0.001 exhibited poor agreement with the SGA in older adults
Note: All models were adjusted by sex and age. who are institutionalized. Malnutrition according to any
Abbreviations: CI, confidence interval; GLIM, Global Leadership Initiative GLIM criteria combination or the MNA‐FF was not
on Malnutrition; HL, Hosmer Lemeshow; MNA‐FF, Mini Nutritional associated with 5‐year mortality. However, higher levels
Assessment–Full Form; OR, odds ratio; SGA, Subjective Global Assessment. of dependence for basic activities of daily living according
to the Katz scale and malnutrition according to the SGA
were independent predictors of mortality in this study. In
sensitivity of only 15.5%, and the assessment of low food conclusion, the prevalence of malnutrition varies accord-
intake was performed by asking participants through ing to the nutrition assessment tool, and the SGA was the
the Canadian Nutrition Screening Tool, which may have best tool for predicting mortality in this sample of older
contributed to different prevalence values when the adults who are institutionalized.
criterion of low food intake was used.20
Malnutrition is known as an important risk factor for A U T H O R C O N TR I B U T I O N S
mortality.25 However, in the present study, malnourished Guilherme Fonseca Graciano and Isabella Ribeiro Souza
individuals according to the GLIM criteria or the contributed to data collection; Guilherme Fonseca
MNA‐FF did not have a higher OR for death. On the Graciano and Bárbara Chaves Santos contributed
other hand, malnutrition according to the SGA was an to statistical analyses; Guilherme Fonseca Graciano,
independent predictor of 5‐year mortality. In the study by Bárbara Chaves Santos, and Lucilene Rezende Anastácio
Sanz‐Paris et al,18 older adults who are institutionalized contributed to the writing of the article; Guilherme
who were severely malnourished according to the GLIM Fonseca Graciano, Bárbara Chaves Santos, Maria Isabel
criteria had higher mortality rates, and older adults with Toulson Davisson Correia, and Lucilene Rezende
 19412452, 0, Downloaded from https://aspenjournals.onlinelibrary.wiley.com/doi/10.1002/ncp.11120 by Instituto Nacional De, Wiley Online Library on [06/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
NUTRITION IN CLINICAL PRACTICE | 7

Anastácio contributed to revision of the article; and all 8. Lino VTS, Pereira SRM, Camacho LAB, Ribeiro Filho ST,
authors contributed to text review. Buksman S. Adaptação transcultural da Escala de Indepen-
dência em Atividades da Vida Diária (Escala de Katz).
Cadernos de Saúde Pública. 2008;24(1):103‐112.
ACKNOWLEDGMENTS
9. Gurney JM, Jelliffe DB. Arm anthropometry in nutritional
The authors acknowledge Fundação de Amparo a
assessment: nomogram for rapid calculation of muscle
Pesquisa do Estado de Minas Gerais (FAPEMIG) for the circumference and cross‐sectional muscle and fat areas. Am
scholarship given to Bárbara Chaves Santos and the project J Clin Nutr. 1973;26(9):912‐915.
(APQ‐02444‐21), and Conselho Nacional de Desenvolvi- 10. Chumlea WC, Guo S, Roche AF, Steinbaugh ML. Prediction of
mento Científico e Tecnológico (CNPq) (303754/2019‐2; body weight for the nonambulatory elderly from anthropome-
405162/2021‐9 and 308037/2023‐5). try. J Am Diet Assoc. 1988;88(5):564‐568.
11. Osterkamp LK. Current perspective on assessment of human
body proportions of relevance to amputees. J Am Diet Assoc.
C O N F LI C T O F I NT E R E S TS S T A TE M E N T
1995;95(2):215‐218.
Guilherme Fonseca Graciano, Bárbara Chaves Santos, 12. Barbosa‐Silva TG, Bielemann RM, Gonzalez MC, Menezes AMB.
Isabella Ribeiro Souza, and Lucilene Rezende Anastácio Prevalence of sarcopenia among community‐dwelling elderly of a
have no conflict of interests to declare. Maria Isabel medium‐sized South American city: results of the COMO VAI?
Toulson Davisson Correia declares the following conflict study: sarcopenia prevalence in a South American city. J Cachexia
of interests: Nestlé Nutrition (consulting on surgical Sarcopenia Muscle. 2016;7(2):136‐143.
material and lectures), Abbott Nutrition (lectures), Danone 13. Universidade Estadual de Campinas. Tabela Brasileira de
Composição de Alimentos. 4th ed. Núcleo de Estudos e
Nutrition (lectures), Fresenius Kabi (lectures and scientific
Pesquisas em Alimentação, Universidade Estadual de Campi-
board), and Baxter Nutrition (lectures).
nas; 2011.
14. Trumbo P, Schlicker S, Yates AA, Poos M. Dietary reference
ORCID intakes for energy, carbohydrate, fiber, fat, fatty acids, choles-
Guilherme F. Graciano http://orcid.org/0000-0002- terol, protein and amino acidos. J Am Diet Assoc. 2002;102(11):
8547-2888 1621‐1630.
Isabella R. Souza http://orcid.org/0000-0003-1721-7129 15. Frisancho AR. Anthropometric Standards for the Assessment of
Maria Isabel T. D. Correia http://orcid.org/0000-0002- Growth and Nutritional Status. University of Michigan
Press; 1990.
3503-4302
16. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new
Lucilene R. Anastácio http://orcid.org/0000-0002- method of classifying prognostic comorbidity in longitudinal
2269-0722 studies: development and validation. J Chronic Dis. 1987;40(5):
Bárbara C. Santos http://orcid.org/0000-0002- 373‐383.
0828-9879 17. Martin Bland J, Altman DG. Statistical methods for assessing
agreement between two methods of clinical measurement.
REFERENCES Lancet. 1986;327(8476):307‐310.
18. Sanz‐Paris A, Fernández MG, Perez‐Nogueras J, et al. Prevalence
1. Volkert D, Beck AM, Cederholm T, et al. ESPEN guideline on
of malnutrition and 1‐year all‐cause mortality in institutionalized
clinical nutrition and hydration in geriatrics. Clin Nutr. 2019;38(1):
elderly comparing different combinations of the GLIM criteria.
10‐47.
JPEN J Parenter Enteral Nutr. 2021;45(6):1164‐1171.
2. Cereda E, Valzolgher L, Pedrolli C. Mini nutritional assess-
19. Henrique JR, Pereira RG, Ferreira ÁRS, et al. Pilot
ment is a good predictor of functional status in institutiona-
study GLIM criteria for categorization of a malnutrition
lised elderly at risk of malnutrition. Clin Nutr. 2008;27(5):
diagnosis of patients undergoing elective gastrointestinal
700‐705.
operations—a pilot study of applicability and validation.
3. Alibhai SMH, Greenwood C, Payette H. An approach to the
Nutrition. 2020;79‐80:110961.
management of unintentional weight loss in elderly people.
20. Allard JP, Keller H, Gramlich L, Jeejeebhoy KN, Laporte M,
Can Med Assoc J. 2005;172(6):773‐780.
Duerksen DR. GLIM criteria has fair sensitivity and specificity
4. Aung KT. Nutritional status of institutionalized elderly.
for diagnosing malnutrition when using SGA as comparator.
Scholars J Appl Med Sci. 2016;4(10):3608‐3611.
Clin Nutr. 2020;39(9):2771‐2777.
5. Guigoz Y. The Mini Nutritional Assessment (MNA) review of 21. Sanz‐París A, Martín‐Palmero A, Gomez‐Candela C, et al. GLIM
the literature—what does it tell us? J Nutr Health Aging. 2006; criteria at hospital admission predict 8‐year all‐cause mortality
10(6):466‐485. in elderly patients with type 2 diabetes mellitus: results
6. Detsky A, Mclaughlin J, Baker J, et al. What is subjective from VIDA study. JPEN J Parenter Enteral Nutr. 2020;44(8):
global assessment of nutritional status? JPEN J Parenter 1492‐1500.
Enteral Nutr. 1987;11(1):8‐13. 22. Zhang X, Tang M, Zhang Q, et al. The GLIM criteria as an
7. Cederholm T, Jensen GL, Correia MITD, et al. GLIM criteria effective tool for nutrition assessment and survival predic-
for the diagnosis of malnutrition—a consensus report from the tion in older adult cancer patients. Clin Nutr. 2021;40(3):
global clinical nutrition community. Clin Nutr. 2019;38(1):1‐9. 1224‐1232.
 19412452, 0, Downloaded from https://aspenjournals.onlinelibrary.wiley.com/doi/10.1002/ncp.11120 by Instituto Nacional De, Wiley Online Library on [06/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 | GRACIANO ET AL.

23. Sperandio N, Priore SE. Inquéritos antropométricos e alimen- SU PPOR T ING INF OR MATI ON
tares na população Brasileira: importante fonte de dados para Additional supporting information can be found online
o desenvolvimento de pesquisas. Cien Saude Colet. 2017;22(2): in the Supporting Information section at the end of this
499‐508.
article.
24. Muñoz Fernandez SS, Garcez FB, Alencar JCG, et al.
Applicability of the GLIM criteria for the diagnosis of
malnutrition in older adults in the emergency ward: a pilot
validation study. Clin Nutr. 2021;40(11):5447‐5456. How to cite this article: Graciano GF, Souza IR,
25. Leiva Badosa E, Badia Tahull M, Virgili Casas N, et al.
Correia MITD, Anastácio LR, Santos BC. Global
Hospital malnutrition screening at admission: malnutrition
Leadership Initiative on Malnutrition criteria in
increases mortality and length of stay. Nutr Hosp. 2017;34(4):
907‐913. older adults who are institutionalized: Agreement
26. Yeung SSY, Chan JHY, Chan RSM, Sham A, Ho SC, Woo J. with the Subjective Global Assessment and its
Predictive value of the GLIM criteria in Chinese community‐ impact on 5‐year mortality. Nutr Clin Pract. 2024;
dwelling and institutionalized older adults aged 70 years and 1‐8. doi:10.1002/ncp.11120
over. J Nutr Health Aging. 2021;25(5):645‐652.