The stability, reliability, and validity of videofluoroscopy measures for patients with head and neck cancer.

The purpose of this study was to contrast the psychometric properties (stability, test-retest reliability, construct, and concurrent validity) of three different tools used for evaluating videofluoroscopy swallowing studies (VFSS): (1) rating the presence or absence of a swallowing disorder, (2) the Bethlehem Assessment Scale (BAS), and (3) biomechanical measures. These three tools were applied to the same three examinations of two different consistencies (liquid and semisolid), taken from 40 VFSSs of patients with head and neck (H&N) cancer. Stability of swallowing across three swallows was a concern for three measures with the liquid consistency and nine measures with the semisolid consistency. Test-retest reliability was found to vary considerably for the two consistencies (liquids, 0.53-1.00; semisolids, 0.45-1.00). Examination of construct validity of the BAS and biomechanical measures indicated that six factors represented swallowing function, but different factors represented swallowing under liquid and semisolid conditions. Concurrent validity of the presence/absence of disorder variables was less than adequate. These results are discussed in the following contexts: (1) psychometric properties of VFSS may not be adequate for clinical and research environments and (2) psychometric properties of VFSS measures appear to vary as a function of bolus consistency.