Adaptation and Validation of the Dynamic Imaging Grade of Swallowing Toxicity for Flexible Endoscopic Evaluation of Swallowing: DIGEST-FEES

Abstract

Purpose While flexible endoscopic evaluation of swallowing (FEES) is a common clinical procedure used in the head and neck cancer (HNC) population, extant outcome measures for FEES such as bolus-level penetration–aspiration and residue scores are not well suited as global patient-level endpoint measures of dysphagia severity in cooperative group trials or clinical outcomes research. The Dynamic Imaging Grade of Swallowing Toxicity (DIGEST) was initially developed and validated for use during videofluoroscopic evaluations as a way to grade safety, efficiency, and overall pharyngeal swallowing impairment. The purpose of this study was to adapt and validate DIGEST for use with FEES. Method A modified Delphi exercise was conducted for content validation, expert consensus, adaptation, and operationalization of DIGEST-FEES. Three blinded, expert raters then evaluated 100 de-identified post-HNC treatment FEES examinations. Intra- and interrater reliability were tested with quadratic weighted kappa. Criterion validity against the MD Anderson Dysphagia Inventory, Functional Oral Intake Scale, Secretion Severity Scale, and Yale Residue Rating Scale was assessed with Spearman correlation coefficients. Results Interrater reliability was almost perfect for overall DIGEST-FEES grade (κ w = 0.83) and safety grade (κ w = 0.86) and substantial for efficiency grade (κ w = 0.74). Intrarater reliability was excellent for all raters (0.9–0.91). Overall DIGEST-FEES grade correlated with MD Anderson Dysphagia Inventory ( r = −.43, p < .0001), Functional Oral Intake Scale ( r = −.43, p < .0001), Secretion Severity Scale ( r = .47, p < .0001), Yale Vallecular Residue ( r = .73, p < .0001), and Yale Pyriform Sinus Residue ( r = .65, p < .0001). Conclusion DIGEST-FEES is a valid and reliable scale to describe the severity of pharyngeal dysphagia in patients with HNC. Supplemental Material https://doi.org/10.23641/asha.14642787