How much do preverbal children signal a wish to be fed? Nested case control study comparing weight faltering and healthy infants.

We aimed to 1/develop an observational tool to rate non-verbal cues infants give when being fed 2/test whether these differ between healthy children and those with weight faltering (WF) 3/describe how well these predict whether offered food is eaten.<bold>Subjects: </bold>and methods: The study used videos of infants eating a standardised meal studied in a case control study nested within the Gateshead Millennium Study (GMS). Infants with weight faltering (WF) were each matched to 2 healthy controls. Half the control videos (N = 28) were used to develop the scale. Food offers were identified and the child's head, eyes, hands, and mouth position/activity rated as signalling a readiness to be fed (engaged), or not (disengaged) as well as whether food was accepted; 5 of these videos were used to assess inter-rater and test-re- test reliability. The scale was then applied to the videos of 28 WF infants (mean age 15.3 months) and 29 remaining controls (mean age 15.8 months) to identify and code all feeding events.<bold>Results: </bold>test-re-test rates varied from 0.89 for events to 0.74 for head; inter-rater reliability varied from 0.78 for hands to 0.67 for mouth. From 2219 observed interactions, 48% showed at least one engaged element, and 73% at least one disengaged; 67% of interactions resulted in food eaten, with no difference between WF and control. Food was eaten after 73% interactions with any engagement, but also in 62% with disengagement.<bold>Conclusions: </bold>Infants were commonly disengaged during meals, but a majority accepted food despite this. Those with weight faltering did not differ compared to healthy controls. [ABSTRACT FROM AUTHOR]