Validity of four nutritional screening tools against subjective global assessment for inpatient adults in a low-middle income country in Asia

Background/objectives Hospital malnutrition is a common problem worldwide. This study aims to assess the validity of widely used nutritional screening tools for hospitalized adults in acute care settings in Ho Chi Minh City, Vietnam. Subjects/methods Participants in this study were 693 adult patients from six general public hospitals, in a multi-center survey undertaken in April and May, 2016. The criterion validity of the Malnutrition Screening Tool (MST), Malnutrition Universal Screening Tool (MUST), Nutrition Risk Screening (NRS-2002) and Mini Nutrition Assessment-Short Form (MNA-SF), modified MST (MST combined with low BMI), and BMI as independent tools were assessed using Subjective Global Assessment (SGA) or low BMI (<18.5 kg/m(2)) as the reference method. Area under curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated using the ROC curve method to determine the validity of screening tools. Results NRS-2002, modified MST (MST + low BMI), MUST, and BMI at 21 kg/m(2) showed moderate/fair validity compared to the reference method (SGA or BMI). MST alone and MNA-SF showed poor validity due to low sensitivity (41.8 and 35.0% for MST and MNA-SF, respectively). Conclusions Based on specificity and sensitivity, the first choice for the most appropriate screening tool for use in Vietnam is the NRS-2002, following by the MST + BMI, MUST, and BMI alone at the cut-off value of 21 kg/m(2). Further investigation on the feasibility and acceptability are required to determine the most appropriate screening tools for use within the Vietnamese context.