ARS | Publication request: Responsiveness to child feeding cues: an observational scale

Submitted to: Obesity
Publication Type: Abstract Only
Publication Acceptance Date: January 3, 2008
Publication Date: October 3, 2008
Citation: Hodges, E., Liu, Y., Johnson, S., Hughes, S.O., Fisher, J.O. Responsiveness to child feeding cues: an observational scale [abstract]. Obesity. 16(Suppl.1):S273.

Technical Abstract: Mismatched caregiver responsiveness to child hunger and satiety cues, is thought to contribute to obesity in infancy and beyond. Assessment of this proposition, however, has been limited by a lack of reliable and valid measures. This research evaluated the interrater reliability of a new observational measure of caregiver responsiveness to feeding cues during infancy and toddlerhood. Interrater reliability of the Responsiveness to Child Feeding Cues Scale (RCFCS) was assessed using cross-sectional observational data from 120 ethnically diverse (White: 30.8%; Black: 35%; Hispanic: 34.2%), mothers and their young children (57.5% female), aged 7 to 24 months. Sessions were recorded in a home-like lab setting from 10 minutes prior to, until 1 minute following the feed. The RCFCS was used to rate caregiver responsiveness to the presence and strength of child hunger cues leading up to the feed (N=20), receptiveness/disinterest in eating at feeding initiation, and fullness cues across the course of the feed (N=28). Feeding cues were categorized as early, active, or late to reflect their typical temporal appearance and intensity. Dimensions of caregiver and infant feeding sensitivity included: visual attentiveness, positive and negative expressiveness, physical disposition, and affect. Intraclass correlation (ICC) and multirater kappa (K) assessed interrater agreement among three raters for 25% (N=30) of the sample. Mean Wt/length Z score was -0.012 (+ 0.89) with 3.4% (N=4) overweight (greater than 95th %; N=118). Average ICCs were 0.99, 0.94, and 1.00 for early, active, and late hunger/receptiveness cues, respectively. Average ICCs were 0.98, 0.96, and 1.00 for early, active, and late disinterest/fullness cues, respectively. Reliabilities (ICC) for maternal responsiveness to child hunger, receptiveness, and fullness cues were 0.90, 0.82, and 0.94, respectively. Reliabilities (ICC or K) for dimensions of feeding sensitivity were greater or equal to 86 %, except for physical disposition. Results suggest that feeding cues and caregiver responsiveness to them from infancy through toddlerhood can be captured reliably by multiple independent raters using the RCFCS.