LAUSR

BACKGROUND Nutrition screening tools are used to identify risk of malnutrition or change in risk of malnutrition. However, it is unclear which tools have demonstrated high validity, reliability, and agreement. OBJECTIVE Our aim was to conduct a systematic review of valid and reliable pediatric nutrition screening tools for identifying malnutrition risk (under- or overnutrition), and to determine whether there are differences in validity and reliability according to users of the tools. METHODS A literature search using Medline, Embase, and CINAHL databases was conducted to identify relevant research published between 1995 and May 2017 examining validity and reliability of nutrition screening tools in the pediatric population. A multidisciplinary workgroup developed eligibility criteria, data were extracted and summarized, risk of bias was assessed, and evidence strength was graded, according to a standard process. RESULTS Twenty-nine studies met inclusion criteria. Thirteen pediatric nutrition screening tools designed for various settings were included in the review (seven inpatient/hospital, three outpatient or specialty setting, and three community). The most frequently examined tools were the Screening Tool for the Assessment of Malnutrition in Pediatrics, Screening Tool for Risk on Nutritional Status and Growth (13 studies each), and Paediatric Yorkhill Malnutrition Score (nine studies). No tools demonstrated high validity. Reliability and agreement were reported infrequently. CONCLUSIONS Nutrition screening tools with good/strong or fair evidence and moderate validity included the Screening Tool for the Assessment of Malnutrition in Pediatrics, Screening Tool for Risk on Nutritional Status and Growth, and Paediatric Yorkhill Malnutrition Score in the inpatient setting and Nutrition Risk Screening Tool for Children and Adolescents with Cystic Fibrosis in the specialty setting. No tools in the community setting met these criteria. While differences in validity and reliability measures among tool users were found, the significance of these findings is unclear. Limitations included few studies examining each tool, heterogeneity between studies examining a common tool, and lack of tools that included currently recommended indicators to identify pediatric malnutrition.