BackgroundAll hospitalized patients should be screened for malnutrition risk. No universal method exists for pediatric patients.MethodsWe performed a cross-sectional study comparing three published malnutrition risk screening tools (PYMS, STAMP, and STRONGkids), applying them to each inpatient aged 1 month to 17 years over a period of five consecutive weekdays in Helsinki University Hospital, Finland.ResultsOf the eligible patients, 67% (n = 69) participated. We found that 6.2% of the children were acutely malnourished and accurately categorized by the three tools. STRONGkids showed the highest specificity (100%) and positive predictive value (36%). Acute malnutrition seemed to be associated with longer hospital stay (p = 0.051).ConclusionSTRONGkids was the most accurate screening tool for detecting acute malnutrition and was therefore chosen as the screening method in our hospital. Routine screening for the risk of malnutrition in pediatric inpatients is important in detecting at-risk children who would otherwise be left without dietary intervention.Electronic supplementary materialThe online version of this article (10.1186/s41043-019-0166-4) contains supplementary material, which is available to authorized users.