Malnutrition in hospitalized patients is a global problem with an increasing prevalence. To date, there is no universally accepted consensus for diagnosing malnutrition. Comparative studies between malnutrition criteria according to GLIM and the gold standard, namely Subjective Global Assessment (SGA) have not been widely carried out in Indonesia. This study aims to compare the validity of the GLIM criteria against the SGA. The study design was cross-sectional with the subject of hospitalized internal medicine patients aged 18-45 years (n=108). The assessment was conducted by comparing prevalence, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), the area under the curve (AUC), positive likelihood ratio (LR+), and negative likelihood ratio (LR-) GLIM criteria against SGA. The results obtained in this study were 57.4 percent women, 60.2 percent late adulthood, 54.6 percent high school graduates, 34.3 percent private employees, and 68.5 percent have income less than minimum wage. The prevalence of malnutrition based on GLIM criteria and SGA was 75 percent and 70.4 percent, respectively. The GLIM criteria have good validity (sensitivity of 98.7%, specificity 81.3%, PPV 92.5%, NPV 96.2%, AUC value 0,9, LR+ 5.28 and a LR- 0.016). Based on the results of the study, it can be concluded that the GLIM criteria have good validity, so they can be recommended as diagnostic tools to determine malnutrition status in hospital nutrition services
Read full abstract