<![CDATA[Introduction. As many as 76.6% of patients aged 60 years and above (elderly) are malnourished or at risk of malnutrition based on MNA within the first 48 hours of hospital admission. The prevalence of malnutrition varies widely depending on the population studied, the healthcare setting and the tools used for the assessment. To date, SGA is widely used in the diagnosis of malnutrition for adults and MNA for elderly patients. Recently, experts proposed empirical consensus of GLIM criteria for determining the diagnosis of malnutrition in adults. This study aimed to investigate the diagnostic performance of GLIM criteria as a new consensus in determining the diagnosis of malnutrition in comparison to MNA as a semi-gold standard for nutritional assessment in the elderly. Methods. A cross-sectional study was conducted from January to April 2022 on 103 consecutively recruited elderly patients aged ≥60 years in the Internal Medicine Wards at Dr. Cipto Mangunkusumo Hospital (RSCM), Jakarta, Indonesia. Results. There were 91.3%, 57.3% and 54.4% malnourished elderly patients upon admission to the hospital, based on GLIM criteria, MNA-SF and MNA-LF, respectively. GLIM has good accuracy compared to both MNA-SF and MNA-LF, as long as the category of malnourished in MNA is a combination of malnutrition and at risk of malnutrition. GLIM had 97.9% sensitivity, 87.5% specificity, AUC 0.93, 98.9% PPV, 77.8% NPV, 7.83 positive LR and 0.02 negative LR towards MNA-SF, and a sensitivity value of 98.9%, specificity 88.9%, AUC 0.94, PPV 98.9%, NPV 88.9%, positive LR 8.91, and negative LR 0.01 towards MNA-LF. Conclusion. GLIM showed good diagnostic accuracy to determine nutritional status in the elderly, especially upon admission to the hospital, so that appropriate early nutritional interventions can be given. Keywords: diagnostic performance, elderly, GLIM, malnutrition, MNA]]>
