Developing a Protocol for Quantitative Analysis of Liquid Swallowing in Children

Abstract

Purpose: Objective measures in videofluoroscopic swallow studies (VFSSs) can quantify swallow biomechanics. There are a wide array of validated measures studied in infants, children, and adults. There is a need for a pediatric VFSS protocol that consists of a small number of vital, time efficient, and clinically relevant measures. In this study, we aimed to establish a standard protocol for quantitative VFSS analysis in children.

Method: Protocol development began with a systematic literature review, which identified 22 quantitative and eight descriptive measures available in the literature. A pediatric VFSS database of 553 children was collected using a standardized VFSS protocol. Studies were evaluated using the 30 previously reported measures covering displacement and timing parameters as well as penetration-aspiration and residue. Measures were tested for rater reliability and internal consistency. Measures meeting acceptable values for protocol inclusion were included in the final protocol (Cronbach's alpha > .53).

Results: Interrater and intrarater reliability of 17 measures met acceptable reliability levels. During internal consistency testing, we removed six further measures based on Cronbach's alpha levels indicating that two or more measures were equivalent in measuring the same aspect of swallow biomechanics in children. A VFSS protocol of reliable, valid, and obtainable objective quantitative (n = 6) and descriptive measures (n = 3) with separate protocols for young infants (≤ 9 months) and older children was established.

Conclusions: A standardized quantitative VFSS protocol for children has been developed to suit two age groups (≤ 9 and > 9 months old). Consistent VFSS administration and reporting support assessment over time and across disease groups. Future research should focus on how this information can be used by clinicians to produce individualized treatment plans for children with swallowing impairment.