ABSTRACT
Nutritional risk screening (NRS) is not yet established in many clinical settings. This study aimed to evaluate the efficacy of two NRS tools; the Paediatric Yorkhill Malnutrition Score (PYMS) and the Screening Tool for the Assessment of Malnutrition in Paediatrics (STAMP), compared to the global dietitians' clinical judgment. The goal of this study was also to estimate the prevalence of nutritional risk in Greek paediatric patients. Overall, 1506 children, 1-16 years, from paediatric and surgical wards of two Greek hospitals were included. NRS was performed using PYMS and STAMP based either on World Health Organization (WHOGC) or Hellenic growth charts (HGC). The first 907 children were also referred to dietitians who categorized children in low, medium and high nutritional risk according to their global clinical judgment. PYMS, either based on WHOGC or HGC, showed better agreement with dietitians' feedback (kPYMS_WHO = 0.47; 95%CI: 0.41-0.52, kPYMS_HGC = 0.48; 95%CI: 0.43-0.53) compared to STAMP (kSTAMP_WHO = 0.28; 95%CI: 0.23-0.33, kSTAMP_HGC = 0.26; 95%CI: 0.21-0.32). PYMS also showed the best diagnostic accuracy compared to STAMP in paediatrics and surgical wards separately. Moreover, the PYMS showed similar sensitivity to the STAMP (WHOGC: 82% vs. 84.4%), but a higher positive predictive value (WHOGC: 58.2 vs. 38.7). Using PYMS, high and medium malnutrition risk was observed at 14.9%, and 13.1% of children, respectively. Almost 28% of hospitalised children were at nutritional risk. Children in hospitals should be screened with effective and feasible NRS tools such as PYMS.
