Reliability for Identification of a Select Set of Temporal and Physiologic Features of Infant Swallows

There is little reported evidence regarding the reliability of temporal and physiologic features of infant swallowing from videofluoroscopic swallowing studies (VFSS). The purpose of this retrospective study was to determine a reliable set of temporal and physiologic features from infant swallowing that can be measured from analysis of VFSS. Temporal and physiologic features for testing were determined from review of previously reported features of infant VFSS in the literature. Two novel analysts underwent three training sessions to learn and practice visual recognition of the proposed features. The two analysts then assessed 25 swallows from 10 total subjects' VFSS. To establish inter- and intra-rater reliability, calculation of Pearson's r was used for features that met criteria for parametric analysis and Spearman's rank correlation coefficient was used for the non-continuous features. Percent agreement was used to report on the reliability of the dichotomous features due to insufficient variability for Spearman's rho analyses. Fifteen of the 16 tested features were found to have acceptable inter- and intra-rater reliability measures, with each analyst achieving a correlation of 0.75 or higher. This project identifies 15 variables that can be reliably measured from infant VFSS. This information can be used to assist with determination of normal versus abnormal swallow features and in developing and testing therapeutic strategies for infants with dysphagia.