A multicentre development and evaluation of a dietetic referral score for nutritional risk in sick infants.

Background & Aims: Unrecognized nutritional issues may delay recovery in hospitalized infants. It has been proposed that nutritional risk screening should be performed at hospital admission, but few tools include infants. The aim of this study was to develop and test a tool to identify sick infants in need of dietetic input.

Methods: Hospitalised infants were recruited from hospitals in the United Kingdom (UK), Greece and Iran. Weight, skinfold thickness and mid upper arm circumference (MUAC) were measured, with detailed dietetic assessment in the UK and Greece. Simple screening questions were used in the UK cohort to formulate a score (infant early nutrition warning score-iNEWS) which was then validated in the Greek and Iranian groups.

Results: After dietetic assessment, 20 (9.6%) UK and 22 (22%) Greek infants were rated as needing dietetic input. Underweight, poor weight gain/loss and reduced intake were all independent predictors of perceived need for dietetic input in stepwise multivariate regression analysis. The score based on these items (iNEWS), had 84% sensitivity, 91% specificity and 49% positive predictive value to predict need for dietetic input in the UK cohort. In the Greek cohort this was 86%, 78% and 53% respectively. In all three countries, infants with high iNEWS had significantly lower average skinfold thickness (between -1 and -1.8 SD, p < 0.0001) and MUAC (between -1.8 and -2 SD, p < 0.0001) than those at low risk.

Conclusions: iNEWS, a simple nutritional risk tool, identifies most hospitalised infants who need dietetic input.

Clinical Trial Registration: https://clinicaltrials.gov/ct2/show/NCT03323957.