Background & aim: This study aims to prospectively assess the validity, reliability and feasibility of two nutrition screening tools NRS-2002 and MST combined with BMI (MST+BMI) after administration by medical doctors in hospitals in Ho Chi Minh City (HCMC) Viet Nam.Methods: Participants were 150 adult patients (validity study), 30 adult patients (reliability study) and 40 medical doctors (feasibility study) in three wards from three general public hospitals in HCMC. Area Under the ROC curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value and percent of correctly classified were calculated to determine validity against SGA or BMI. Cohen's kappa (κ) statistics were calculated to determine the reliability of NRS-2002 and MST+BMI. Viewpoints regarding the benefit and role of nutrition screening and the ease of use, acceptability and feasibility of NRS-2002 and MST+BMI were assessed among medical doctors using self-administered questionnaires with Likert scales to determine feasibility.Results: After completing by medical doctors the NRS-2002 and MST+BMI showed good validity (the sensitivity and specificity were 80.3% and 79.8% for NRS-2002 and were 81.8% and 84.5% for MST+BMI, respectively) and consistent reliability (κ: 0.72 & 0.75) against SGA or BMI. Medical doctors mostly agreed (93%) on the need for and benefits of nutrition screening. It took between two and 4 min to complete screening using NRS-2002 or MST+BMI. Over 82% of medical doctors indicated that the NRS-2002 and MST+BMI were easy to conduct. There was, however, lower agreement (73%), regarding the feasibility of implementing screening in their wards and hospitals.Conclusions: Both NRS-2002 and MST+BMI were valid, reliable and feasible for use by medical doctors for nutrition screening in hospitals in the resource sparse Vietnamese context. Additional activities are required to make nutrition screening more feasible.
Keywords: MST; NRS-2002; SGA; Validity; feasibility; medical doctor; reliability.