Diagnostic performance of mini nutritional Assessment-Short form against the global leadership initiative on malnutrition criteria Authors: Aleksandra Kaluźniak-Szymanowska, Ewa Deskur-Śmielecka, Roma Krzymińska-Siemaszko, Arkadiusz Styszyński, Sławomir Tobis, Marta Lewandowicz-Czarnecka, Jerzy Chudek, Tomasz Kostka, Małgorzata Mossakowska, Karolina Piotrowicz, Hanna Kujawska-Danecka, Katarzyna Wieczorowska-Tobis Published: 2025 ABSTRACT AbstractIn 2018, the Global Leadership Initiative on Malnutrition (GLIM) experts accepted screening for malnutrition with any validated tool. In gerontological dietetics, Mini Nutritional Assessment-Short Form (MNA-SF) remains the most frequently used one. We evaluated the diagnostic performance of MNA-SF against the GLIM criteria in a representative group of older adults (PolSenior2 study population). The complete dataset was available for 5614 participants. We compared two groups of respondents: fulfilling GLIM criteria (GLIM malnutrition, positive screening) and GLIM malnutrition, screening negative. At least one phenotypic and one etiological GLIM criterion was diagnosed in 1136 (20.2%) study participants, including 789 with positive screening results. Therefore, malnutrition would have been undiagnosed in one-third of the malnourished group (30.5% of the group was diagnosed based on the GLIM criterion). In comparison with participants with GLIM malnutrition, screening positive, those with GLIM malnutrition, screening negative were younger and more often had an acute or chronic inflammatory disease. They were less likely to report unintentional weight loss and reduced food intake, and they had higher body weight and BMI. The sensitivity of the MNA-SF in diagnosing malnutrition was estimated at 69.5%. Based on our findings, using a high clinical malnutrition risk profile as an alternative to screening tools for malnutrition in the GLIM diagnostic algorithm could be beneficial.Keywords: Malnutrition, GLIM, MNA-SF, Diagnostic performanceSubject terms: Health care, Medical research, Risk factors Introduction Numerous factors, including age-related changes, multimorbidity, polypharmacy, and socioeconomic conditions, make older people particularly vulnerable to deterioration in nutritional status1–3. Malnutrition, in turn, has significant negative consequences in older people, including longer and more frequent hospitalizations, loss of functional efficiency worsening overall health status, and increased risk of death2,4,5. Despite its recognized adverse health outcomes, malnutrition is rarely diagnosed in daily clinical practice6. The lack of consistent diagnostic criteria is usually blamed for this situation7. In 2018, experts from the European Society of Clinical Nutrition and Metabolism (ESPEN), the American Society of Parenteral and Enteral Nutrition (ASPEN), the Latin American Federation of Parenteral and Enteral Nutrition (FELANPE), and the Parenteral and Enteral Nutrition Society of Asia (PENSA) assembled in the Global Leadership Initiative on Malnutrition (GLIM) group developed and presented consistent criteria of malnutrition8. The GLIM diagnostic path has two stages. The first stage involves a screening test using any validated tool [e.g., Mini Nutritional Assessment-Short Form (MNA-SF), Nutritional Risk Score (NRS-2002), Malnutrition Universal Screening Tool (MUST)]. Currently, MNA-SF is one of the most frequently used questionnaires in gerontological dietetics6,9. According to GLIM algorithm, all individuals with a positive screening result proceed to the second stage, consisting of an assessment of three phenotypic criteria [involuntary weight loss, low body mass index (BMI), and reduced muscle mass] and two etiological criteria (acute or chronic inflammatory disease and reduction in food intake or impaired food digestion or absorption). Confirmation of at least one criterion from each group allows for the diagnosis of malnutrition. However, the efficacy of the screening tools has been called into question10–14. For example, our earlier study enrolling 273 community-dwelling older adults (mean age 72.1±7.7 years) demonstrated that performing the full diagnostics of malnutrition only in those with a positive screening result (MNA-SF) led to missed diagnoses in as many as 40% of people with the condition. These individuals met one phenotypic criterion and one etiological criterion, thus allowing for the diagnosis of malnutrition despite the lack of a positive screening result15. To the best of our knowledge, there is a lack of population-based studies assessing the diagnostic performance of the screening tools against the GLIM criteria. Therefore, we evaluated the diagnostic performance of MNA-SF against the GLIM criteria in a representative cohort of the older adult population in Poland. Methods The PolSenior2 project was conducted between 2018 and 2019 on a representative sample of the older population in Poland living in both rural and urban areas. The aim of the PolSenior2 was to conduct a comprehensive assessment of the elderly community-dwelling Polish population, encompassing medical, psychological, social, and economic aspects. Respondents were selected through a multistage random sampling process, independently for men and women across seven 5-year age cohorts (60–65, 65–69, 70–74, 75–79, 80–84, 85–89, and 90 + years), with an intentional over-representation of men and the oldest participants compared to the actual structure of the Polish population. The over-representation enabled comparisons between sexes and age groups. Finally, 5987 respondents aged 60 and older were enrolled in the PolSenior2 study. The study protocol consisted of three home visits by specially trained nurses. A comprehensive geriatric assessment was performed during the first visit, including evaluating nutritional status with the MNA-SF questionnaire. If there were issues with the respondent’s ability to answer, the responses were provided by a caregiver. The interviewers noted each such situation. The study protocol was approved by the Bioethics Committee of the Medical University of Gdansk (NKBBN/257/2017). All procedures were performed in accordance with applicable national and international ethical standards and regulations, including the principles outlined in the Declaration of Helsinki. All respondents or their caregivers gave informed consent before participation. The study details have been described elsewhere16. Data on individuals who completed the MNA-SF questionnaire and were eligible for assessment of phenotypic and etiological malnutrition criteria were included in the analysis. Participants for whom clear-cut confirmation or exclusion of malnutrition was impossible due to missing data were excluded from the analysis. The complete data necessary to perform the nutritional status assessment with the GLIM criteria were available in 5614 participants (2864 women and 2750 men; 93.8% of the PolSenior2 population). Nutritional status assessment. Nutritional status assessment. The GLIM diagnostic criteria were used to assess nutritional status of all study participants. The MNA-SF questionnaire was used as a screening tool. This questionnaire was developed to assess the nutritional status of older adults17 and was validated in the Polish population18. The questionnaire asks about decreased food intake, weight loss, and psychological stress or acute disease in the past three months. The remaining items refer to neuropsychological problems, impaired mobility, and low BMI. Scores ranging from 12 to 14 points indicate normal nutritional status, while scores of 11 points or less indicate screen- positive17, necessitating further diagnostics according to the GLIM algorithm. To evaluate the diagnostic performance of the MNA-SF questionnaire, we assessed the phenotypic and etiological criteria in all participants, regardless of their screening results. Malnutrition was diagnosed in individuals who met at least one criterion from each group. We use the words ‘malnutrition’ and ‘undernutrition’ synonymously. 1.1 Phenotypic criteria: The weight loss criterion was met if a respondent reported an unintentional loss of > 5% of usual body weight in the last six months, The weight loss criterion was met if a respondent reported an unintentional loss of > 5% of usual body weight in the last six months, Low BMI was defined as < 20 kg/m² for individuals up to 70 years and < 22 kg/m² for those aged 70 and over, Low BMI was defined as < 20 kg/m² for individuals up to 70 years and < 22 kg/m² for those aged 70 and over, Low muscle mass (LMM) was assessed based on moderately low calf circumference, which should be assessed in older adults, using the following cutoff points: ≤33 cm for women and ≤ 34 cm for men19,20. Low muscle mass (LMM) was assessed based on moderately low calf circumference, which should be assessed in older adults, using the following cutoff points: ≤33 cm for women and ≤ 34 cm for men19,20. 1.2 Etiological criteria: Reduced food intake was recognized in individuals reporting any decline in meal intake in the past three months or reduced assimilation (presence of peptic ulcer disease), Reduced food intake was recognized in individuals reporting any decline in meal intake in the past three months or reduced assimilation (presence of peptic ulcer disease), The presence of acute or chronic inflammatory disease was stated based on information from interviews with participants (chronic obstructive pulmonary disease, liver cirrhosis, cancer, stroke) or positive blood inflammatory marker [C-reactive protein (CRP) ≥ 5 mg/L). The presence of acute or chronic inflammatory disease was stated based on information from interviews with participants (chronic obstructive pulmonary disease, liver cirrhosis, cancer, stroke) or positive blood inflammatory marker [C-reactive protein (CRP) ≥ 5 mg/L). We compared two groups of patients: participants with malnutrition diagnosed based on the presence of at least one phenotypic and at least one etiological criterion (GLIM malnutrition, screening negative) and those meeting the full GLIM criteria (GLIM malnutrition, screening positive), i.e., a positive screening result (MNA-SF results 11 points or less) and the co-occurrence of an etiological and phenotypic criterion. The study flow is presented on Fig. 1. Study flow. MNA-SF: Mini Nutritional Assessment- Short Form; GLIM: Global Leadership Initiative on Malnutrition. 2.Statistical analysis. Statistical analysis. Quantitative variables were presented as mean and standard deviation (SD), while qualitative variables were presented as absolute (n) and relative (%) values. The normal distribution of all quantitative variables was verified with the Shapiro-Wilk test. The homogeneity of variances was checked using Levene’s test. For quantitative variables, the differences between groups were assessed with Student’s t-test (normal distribution, homogeneity of variances), Cochrane-Cox test (normal distribution, lack of homogeneity of variances), or Mann-Whitney U test (lack of normal distribution). The chi-square test with Yates’ correction was used to compare qualitative variables between the two groups. To evaluate the diagnostic performance of the MNA-SF against the GLIM criteria, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), area under the ROC curve (AUC), and kappa coefficient were calculated. Sensitivity and specificity were classified as follows: good - >80%, fair − 50–80%, or poor - <50%21. The diagnostic accuracy of the screening test was characterized by the AUC: good - >0.8, fair − 0.6–0.8, or poor accuracy - <0.621. The kappa coefficient was used to assess the agreement between the MNA-SF and GLIM results: very good - >0.8, good − 0.61–0.8, moderate − 0.41–0.6, fair − 0.21–0.4, or poor agreement - <0.221. A p-value of < 0.05 was considered significant. Statistical calculations were performed using the STATISTICA 10 PL software package. Results Prevalence of malnutrition. Prevalence of malnutrition. Table 1 shows the prevalence of particular phenotypic and etiological GLIM criteria and the results of the nutritional status screening with the MNA-SF questionnaire. Based on the MNA-SF score, nearly one-third of the study population (n = 1568, 27.9%) had impaired nutritional status (malnutrition or risk of malnutrition). Half of them (n = 789, 50.3%) met at least one phenotypic and etiological GLIM criterion concomitantly (Fig. 2), which accounted for 14.1% of the study sample. If GLIM criteria were applied only in individuals with a positive screening result (in line with GLIM algorithm), malnutrition would have been undiagnosed in one-third of the malnourished group (n = 347, 30.5%), that is 6.2% of the whole study group. At least one phenotypic criterion was found in over one-third of the participants (n = 1925, 34.3%). Low muscle mass was the most commonly observed phenotypic criterion (n = 1075, 19.7%). Unintentional weight loss was reported by nearly one in five individuals (n = 982, 17.6%), whereas a low BMI was identified in only 6.2% (n = 344) of participants. Table 1Prevalence of phenotypic and etiological criteria in the study group divided by sex.VariableSexTotaln (%)Womenn (%)Menn (%)p-valuePhenotypic criteriaUnintentional weight loss(n = 5591)no4609 (82.4)2340 (82.0)2269 (82.9)0.403yes982 (17.6)513 (18.0)469 (17.1)BMI(n = 5540)normal5196 (93.8)2627 (93.2)2569 (94.4)0.059decreased344 (6.2)192 (6.8)152 (5.6)Calf circumference (low muscle mass)(n = 5464)normal4389 (80.3)2262 (80.8)2127 (79.8)0.324decreased1075 (19.7)536 (19.2)539 (20.2)At least one phenotypic criterion (n = 5614)no3689 (65.7)1886 (65.9)1803 (65.6)0.820yes1925 (34.3)978 (34.1)947 (34.4)Etiological criteriaDecreased food intake or assmilation(n = 5614)no4326 (77.1)2194 (76.6)2132 (74.5)0.430yes1288 (22.9)670 (23.4)618 (22.5)Acute or chronic inflammation disease or elevated levels of inflammatory markers(n = 5614)no3523 (62.8)1832 (64.0)1691 (61.5)0.059yes2091 (37.2)1032 (36.0)1059 (38.5)At least one etiological criterion (n = 5614)no2808 (50.0)1435 (50.1)1373 (49.9)0.915yes2806 (50.0)1429 (49.9)13,777 (50.1)MNA-SFnormal nutritional status4046 (72.1)1978 (69.1)2068 (75.2)< 0.001malnutrition or risk of malnutrition1568 (27.9)886 (30.9)682 (24.8)Values are presented as numbers (%).BMI, Body Mass Index; MNA-SF, Mini Nutritional Assessment-Short Form. Prevalence of phenotypic and etiological criteria in the study group divided by sex. Unintentional weight loss (n = 5591) BMI (n = 5540) Calf circumference (low muscle mass) (n = 5464) Decreased food intake or assmilation (n = 5614) Acute or chronic inflammation disease or elevated levels of inflammatory markers (n = 5614) Values are presented as numbers (%). BMI, Body Mass Index; MNA-SF, Mini Nutritional Assessment-Short Form. Participants with GLIM malnutrition, screening negative, participants with impaired nutritional status based on the MNA-SF score (MNA-SF), and individuals with GLIM malnutrition, screening positive. MNA-SF, Mini Nutritional Assessment- Short Form; GLIM, Global Leadership Initiative on Malnutrition. At least one etiological criterion was observed in half of the study group (n = 2806, 50.0%). Acute or chronic inflammation disease or elevated levels of inflammatory markers were the most common, found in nearly two out of five participants (n = 2091, 37.2%). Decreased food intake was the least frequent criterion, applying to more one-fifth of individuals (n = 1288, 22.9%). The co-occurrence of at least one phenotypic and one etiological criterion was found in one-fifth of the participants (n = 1136, 20.2%). These respondents were considered malnourished (GLIM malnutrition, screening negative) in further analysis. 2.Malnutrition in different diagnostic models. Malnutrition in different diagnostic models. Table 2 details the characteristics of individuals identified as malnourished according to two proposed diagnostic approaches. In comparison with participants with GLIM malnutrition, screening positive those with GLIM Malnutrition, screening negative were younger (p < 0.001), and more often had an acute or chronic inflammatory disease (p < 0.001), despite lower CRP levels (p = 0.006). They were less likely to report unintentional weight loss (p < 0.001) and reduced food intake (p < 0.001), and they had higher body weight and BMI (p < 0.001 and p = 0.013, respectively). The difference approached statistical significancewere observed in calf circumference between the groups (p = 0.048). Table 2Characteristics of individuals with malnutrition diagnosed with different diagnostic models.TotalGLIM malnutrition, screening negativeGLIM malnutrition, screening positivep- valuen (%)M ± SDn (%)M ± SDn (%)M ± SDSex Men576 (50.7)205 (59.1)371 (47.0)< 0.001 Women560 (49.3)142 (40.9)418 (53.0)Age (years) 60–64114 (10.0)47 (13.5)67 (8.5)< 0.001 65–69145 (12.8)49 (14.1)96 (12.2) 70–74171 (15.1)66 (19.0)105 (13.3) 75–79180 (15.8)60 (17.3)120 (15.2) 80–84184 (16.2)60 (17.3)124 (15.7) 85–89175 (15.4)41 (11.8)134 (17.0) 90 and over167 (14.7)24 (7.0)143 (18.2)Acute or chronic inflammatory disease Yes713 (65.2)288 (83.0)509 (64.5)< 0.001 No380 (34.8)59 (17.0)280 (35.5)Unintentional weight loss Yes651 (57.5)125 (36.1)526 (66.8)< 0.001 No482 (42.5)221 (63.9)261 (33.2)Reduced food intake Yes653 (57.5)104 (30.0)549 (69.6)< 0.001 No483 (42.5)243 (70.0)240 (30.4)Weight (kg) n = 109371.2 ± 16.273.4 ± 13.870.2 ± 17.1< 0.001Height (cm) n = 1135163.7 ± 9.4164.7 ± 8.8163.7 ± 9.70.016BMI n = 109326.5 ± 5.227.0 ± 4.226.2 ± 5.60.013MNA-SF (points) n = 113610.2 ± 2.513.0 ± 0.89.0 ± 1.9< 0.001CRP n = 11048.6 ± 17.36.8 ± 11.79.4 ± 19.30.006Calf circumference (cm) n = 111034.4 ± 5.434.0 ± 4.534.6 ± 5.80.048Values are presented as numbers (%) or * mean ± standard deviation.BMI, Body Mass Index; MNA-SF, Mini Nutritional Assessment-Short Form; CRP, C-reactive protein; GLIM, Global Leadership Initiative on Malnutrition. Characteristics of individuals with malnutrition diagnosed with different diagnostic models. Calf circumference (cm) n = 1110 Values are presented as numbers (%) or * mean ± standard deviation. BMI, Body Mass Index; MNA-SF, Mini Nutritional Assessment-Short Form; CRP, C-reactive protein; GLIM, Global Leadership Initiative on Malnutrition. Table 3 presents the differences between the GLIM malnutrition, screening negative group, and the MNA-SF positive (GLIM negative) group. Individuals with a negative screening test result who met at least one phenotypic criterion and at least one etiologic criterion were more often male (p < 0.001). In this group, chronic or intermittent inflammatory conditions were more prevalent (< 0.001), and elevated CRP levels were observed (< 0.001). Participants with a GLIM malnutrition, screening negative had lower BMI and calf circumference (< 0.001). Table 3Characteristics of individuals with GLIM malnutrition, screening negative compared to individuals with GLIM negative, MNA-SF positive.GLIM malnutrition, screening negativeMNA-SF risk of malnutritionp- valuen (%)M ± SDn (%)M ± SDSex Men205 (59.1)311 (39.9)< 0.001 Women142 (40.0468 (60.1)Age (years) 60–6447 (13.5)86 (11.0)< 0.001 65–6949 (14.1)102 (13.1) 70–7466 (19.0)106 (13.6) 75–7960 (17.3)107 (13.7) 80–8460 (17.3)135 (17.3) 85–8941 (11.8)123 (15.8) 90 and over24 (7.0)120 (15.4)Acute or chronic inflammatory disease Yes288 (83.0)251 (32.2)< 0.001 No59 (17.0)528 (67.8)Unintentional weight loss Yes125 (36.1)143 (18.5)< 0.001 No221 (63.9)632 (81.5)Reduced food intake Yes38 (11.0)189 (23.2)< 0.001 No309 (89.0)598 (76.8)Weight (kg)73.4 ± 13.875.6 ± 16.90.037Height (cm)164.7 ± 8.8162.7 ± 9.2< 0.001BMI n = 109327.0 ± 4.228.5 ± 5.9< 0.001CRP6.8 ± 11.74.1 ± 9.3< 0.001Calf circumference (cm) 33.9 ± 4.537.5 ± 4.9< 0.001Values are presented as numbers (%) or * mean ± standard deviation.BMI, Body Mass Index; MNA-SF, Mini Nutritional Assessment-Short Form; CRP, C-reactive protein; GLIM, Global Leadership Initiative on Malnutrition. Characteristics of individuals with GLIM malnutrition, screening negative compared to individuals with GLIM negative, MNA-SF positive. Values are presented as numbers (%) or * mean ± standard deviation. BMI, Body Mass Index; MNA-SF, Mini Nutritional Assessment-Short Form; CRP, C-reactive protein; GLIM, Global Leadership Initiative on Malnutrition. 3.Sensitivity, specificity, accuracy and diagnostic value of MNA-SF. Sensitivity, specificity, accuracy and diagnostic value of MNA-SF. The ROC curve for MNA-SF against the GLIM criteria. Discussion The GLIM experts recommend screening for malnutrition with any validated tool as a first step of the diagnostic procedure. Our goal was to assess the diagnostic performance of the commonly used MNA-SF questionnaire, a screening tool for older adults. Screening tools should be characterized with high specificity and AUC and, first of all, high sensitivity. Our analysis showed that MNA-SF had good specificity (82.6%) and accuracy (AUC 0.81), but its sensitivity was only fair (69.5%). These results indicate that this questionnaire has fair diagnostic value as a screening tool for malnutrition. Almost one-third of all negative MNA-SF results were false. Nearly one-fifth of the study participants had GLIM malnutrition, screening negative, and nearly one-third had abnormal nutritional status by MNA-SF screening. However, the results of the two diagnostic approaches were consistent in only one out of seven respondents. Previous studies, including our own, have shown that the sensitivity and specificity of the MNA-SF were only fair (59.2% and 78.8%, respectively)15. Other authors have also demonstrated unsatisfactory diagnostic performance of the MNA-SF as a malnutrition diagnostic tool among community-dwelling older adults. For example, Ozer et al.22 found only moderate agreement between MNA-SF and GLIM criteria (kappa coefficient 0.47) in a study involving 252 older participants (mean age 68 years). The sensitivity and specificity of the MNA-SF for identifying mild and moderate malnutrition were 50% and 80%, while for severe malnutrition, they came to 89% and 85%, respectively22. In the analysis by Sanchez-Rodriguez et al.23, including data from 411 older individuals (mean age 73.2 ± 6.05), the MNA-SF missed the diagnosis of abnormal nutritional status in as many as 60.4% of participants with malnutrition based on the phenotypic and etiological GLIM criteria23. The diagnostic performance against the GLIM criteria of some other malnutrition screening tools have been assessed in previous studies. Van Dronkelaar et al.24 assessed the diagnostic performance of the Short Nutritional Assessment Questionnaire (SNAQ), Malnutrition Universal Screening Tool (MUST), Malnutrition Screening Tool (MST), MNA-SF and the Patient-Generated Subjective Global Assessment-Short Form (PG-SGA-SF) in a sample of 356 Dutch in-hospital patients [median age 70, interquartile range (IQR) 63–77]. None of the assessed questionnaires had high sensitivity24. Bellanti et al.10 in a study involving 152 elderly persons (mean age 77.8 ± 7.8 years), found that the Subjective Global Assessment (SGA) was the only questionnaire with high sensitivity (96%), in contrast to NRS-2002 (sensitivity 47%) and MUST (64%)10. Another study by Garcia-Fuente et al.25 compared five questionnaires: MNA-SF, MST, MUST, NRS-2002 and Nutritional Control Scale (CONUT) as well as three variants (consisting of 3, 8 and 14 questions) of the Seniors in the Community: Risk Evaluation for Eating Nutrition (SCREEN) scale in a sample of 85 hospitalized elderly persons (median age 83, QR 14 years). The highest sensitivity was observed for SCREEN 3 (93%), and the best specificity was achieved for MUST (91%). The diagnostic specificity of MNA-SF was only 68%, which is similar to the results of our study25. Finally, Selcuk et al.26 performed an analysis in 669 older in-patients (mean age 75.5 ± 7.7 years) and found fair to good sensitivity for the MST, MUST, SNAQ, MNA-SF, NRS-2002, and Graz Malnutrition Screening (GMS) questionnaires26. It should be emphasized that the results of the above-mentioned studies, all of which were performed in hospitalized patients, should be compared with caution to our analysis, which included community-dwelling persons. In our study, we observed that the MNA-SF questionnaire identified malnutrition significantly more often in women, which is consistent with findings reported in the available literature27–29. However, it is worth noting that malnutrition diagnosed solely based on the phenotypic and etiologic GLIM criteria (without prior screening) was more frequently identified in men. The underlying cause of this discrepancy remains unclear due to the lack of relevant literature. In our study sample, respondents with false negative results of the MNA-SF were younger, less frequently reported unintentional weight loss or reduced food intake, and had higher weight and BMI. Literature suggests that MNA-SF criteria carry significant risks. For instance, Yeung et al.30 analyzed the association of various GLIM diagnostic criteria with the risk of death within three years among 1576 community-dwelling and 427 institutionalized older adults (mean age 78.1±6.5 years). They observed that unintentional weight loss was significantly associated with a higher risk of death in community-dwelling individuals, and they pointed out that reduced food intake could have contributed to the reduction of body weight30. Similar findings were reported by Kiss et al.31, who analyzed the prevalence of malnutrition using 21 possible combinations of GLIM etiological and phenotypic criteria. Next, they determined combinations that best predicted the risk of death or unplanned hospitalization within 30 days. About half of their study sample were older oncological patients. Kiss et al. observed that a combination of unintentional weight loss, low muscle mass, and reduced food intake was predictive of death. In contrast, the co-occurrence of reduced food intake and weight loss or low muscle mass was associated with a higher risk of unplanned hospitalizations31. Therefore, a diagnostic approach based exclusively on etiological and phenotypic criteria enables recognition of milder forms of malnutrition. Of note, participants with malnutrition diagnosed with this approach were more likely to have an acute or chronic inflammatory disease. This finding suggests that the diagnostic approach based exclusively on the phenotypic and etiological GLIM criteria (without screening test) may detect disease-associated malnutrition (protein-energy malnutrition, PEM). In older adults, PEM is particularly important, as it can exacerbate age-related loss of muscle mass and strength and lead to sarcopenic obesity, which itself should be considered a form of qualitative malnutrition. Both conditions are associated with increased risk of falls, fractures, dependency, and finally functional disability and deterioration of quality of life. It should be emphasized that PEM not only reflects nutritional status but also determines general health status and prognosis in older persons. Hence, the shortening of the diagnostic algorithm for malnutrition, by omitting the screening test in those with high clinical risk of malnutrition, may contribute to earlier recognition of this important health problem of older age and thus accelerate the implementation of a nutritional intervention. Limitation and strength Our study has some limitations, including using calf circumference as a surrogate parameter for muscle mass. Due to the equipment limitations in the PolSenior2 study (inability to calculate the appendicular lean mass- ALM index), we could not use the bioimpedance results for our analysis. Moreover, due to the population-based nature of the study, we considered all cases of peptic ulcer disease and stroke as etiologic criteria; however, future studies should include a more detailed case analysis. Additionally, we could not assess food intake with objective methods, such as 24-hour recall. Instead, we used the subjective respondents’ answers to MNA-SF items referring to food decline over the past three months. Moreover, the results of calf circumference measurements could be biased because multiple persons took them. However, all interviewers in the PolSenior2 study were nurses specially trained for this procedure. The cross-sectional and retrospective nature of the study prevents us from making causal inferences. Our work also has significant strengths. To our knowledge, it was the first study evaluating the diagnostic performance of MNA-SF outside the GLIM diagnostic algorithm in a large population group of community-dwelling older adults. We showed that individuals with malnutrition diagnosed solely with the GLIM criteria often reported reduced food intake and unintentional weight loss. As these factors are common among older adults, healthcare providers should pay special attention to them to prevent malnutrition in geriatric patients. Conclusion We assessed the diagnostic performance of the MNA-SF against GLIM criteria in the diagnostics of malnutrition in a sample representative of the older Polish population. The fair sensitivity of the questionnaire calls into question its use in the first stage of the GLIM algorithm and supports our earlier findings15. Based on our findings, using high clinical risk of malnutrition as an alternative to screening tools of malnutrition in the GLIM diagnostic algorithm could be beneficial. Participants with malnutrition diagnosed based on phenotypic and etiological criteria with omission of the screening phase were more likely to have acute or chronic inflammatory disease, which may be considered as a high clinical risk of malnutrition. Further, population-based studies are needed to confirm our findings and support our proposal to introduce high clinical risk of malnutrition as an alternative to screening tests in malnutrition diagnostic algorithms. Moreover, further studies should assess the prognosis in patients with malnutrition diagnosed based on different criteria. Data Availability A minimal, anonymized database is now available on the PolSenior2 project website at the link: https://polsenior2.gumed.edu.pl/68965.html under the name “Health status correlates of malnutrition”. Author contributions **Aleksandra Kaluźniak-Szymanowska: ** Conceptualization, Methodology, Formal analysis, Data curation, Writing- original draft, Visualization, Project administration; **Ewa Deskur-Śmielecka: ** Formal analysis, Writing- original draft, Visualization; **Roma Krzymińska-Siemaszko: ** Formal analysis, Writing- Reviewing and Editing; **Arkadiusz Styszyński: ** Formal analysis, Writing- Reviewing and Editing; **Sławomir Tobis.:** Formal analysis, Writing- Reviewing and Editing; **Marta Lewandowicz.:** Formal analysis, Writing- Reviewing and Editing; **Jerzy Chudek.:** Conceptualization, Methodology, Data curation, Writing- Reviewing and Editing; **Tomasz Kostka.:** Investigation, Writing- Reviewing and Editing; **Małgorzata Mossakowska.:** Methodology, Data curation, Investigation, Writing- Reviewing and Editing; **Karolina Piotrowicz.:** Investigation, Writing- Reviewing and Editing; **Hanna Kujawska-Danecka: ** Investigation, Writing- Reviewing and Editing; **Katarzyna Wieczorowska-Tobis: ** Conceptualization, Methodology, Writing- Reviewing and Editing, Supervision. Funding This paper was implemented under contract No. 6/5/4.2/NPZ/2017/1203/1257 for the implementation of the task in the field of public health of the Operational Objective No. 5 points 4.2. of the National Health Program for years 2016–2020, entitled “Health Status and Its Socio-economic Covariates of the Older Population in Poland-the Nationwide PolSenior2 Survey” (PolSenior2). Data availability A minimal, anonymized database is now available on the PolSenior2 project website at the link: https://polsenior2.gumed.edu.pl/68965.html under the name “Health status correlates of malnutrition”. Declarations Competing Interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement A minimal, anonymized database is now available on the PolSenior2 project website at the link: https://polsenior2.gumed.edu.pl/68965.html under the name “Health status correlates of malnutrition”. A minimal, anonymized database is now available on the PolSenior2 project website at the link: https://polsenior2.gumed.edu.pl/68965.html under the name “Health status correlates of malnutrition”.