First Steps Towards Development of an Instrument for the Reproducible Quantification of Oropharyngeal Swallow Physiology in Bottle-Fed Children.

The study tested rater agreement, not whether the tool improves diagnosis, feeding decisions, or health outcomes. Only 7 raters scored 10 examinations, and the paper describes the work as a feasibility study. The gold standard was set by the tool developers by consensus, not by an independent reference standard. Agreement allowed one level worse than the gold standard for most components, and up to two extra sucks for component 3, so a score could overcall impairment and still count as reliable. Some components did not reach the reliability criterion and needed additional training. Components most difficult to score included oral containment and clearance, pharyngeal containment and clearance, and airway protection. Airway protection was hard to score because small amounts of barium can be difficult to see and children's airway structures are not fully calcified. Nipple, bottle type, positioning, and screen resolution were not controlled. The tool was tested only with bottle-fed children aged 1 week to 23 months. Construct and external validity were planned for future work, not tested here.

The paper reports research grant funding from NIH/NIDCD.

Do not read the 80% agreement as proof that the tool is clinically valid or that it will improve care. The study only showed that seven raters could be trained to match a developer consensus score on ten examinations.