Examining the discriminant validity of an observational coding system of child behaviour during feeding

Background: Parents of weight faltering children have been found to be less sensitive than control group mothers, to their child’s communication during mealtimes. However, little is known about whether the communication signals the child gives during the meal differ. This study aims to establish discriminant validity of a new observational tool, the Child’s Interactive Mealtime Behaviour (CIMB) scale, designed to measure the child’s communication during meals. Methods: Study one: Mealtime videos were used from 30 children identified as weight faltering and 29 controls identified previously in a nested case control study within the Gateshead Millennium Baby Study (GMBS). Videos were rated using the Child’s Interactive Mealtime Behaviour (CIMB) scale, designed to identify the cues children give their parent regarding their readiness to be fed. Study Two: The CIMB scale was applied to the mealtime videos of a group of 12 children who were outpatients at a hospital feeding clinic. The results from this clinical sample are compared to the control group from the GMBS. Results: No significant difference in feeding behaviour was found between the cases and control group in the GMBS sample. When interactions were analysed individually, irrespective of group, the position of the child’s head and mouth significantly predicted whether the child would eat, but most mouthfuls of food taken were not actively cued by the child. Conclusions: The CIMB scale does not discriminate between children growing normally and children with weight faltering or feeding behaviour problems. It may be that a more global measure is required to identify maladaptive behaviours that discriminate between the groups. The findings indicate that even children with feeding behaviour problems are generally passive during meals and frequently eat even when they do not appear oriented towards the food