Reproducibility of swallowing over time in healthy elderly adults.

The purpose of this study was to evaluate the reproducibility of aspects of swallowing assessed over time by the Exeter Dysphagia Assessment Technique. Fourteen healthy elderly adults were assessed during the morning on four consecutive days. During day 1, the assessment was repeated ten times. During days 2, 3, and 4 the assessment was carried out once and at the same time of the day. This allowed the reproducibility of the data to be examined between sessions and between days. The results show that the reproducibility of data between sessions was moderate to very good for all of the items evaluated (median kappas = 0.56-0.82, ICCs = 0.7-0.8). Between-day reproducibility was also moderate to very good for nine of the variables measured (median kappas = 0.58-0.96, ICCs = 0.53-0.84) and poor for one (ICC = 0.15). We conclude that the majority of variables examined by the Exeter Dysphagia Assessment Technique are reliable when swallowing is assessed repeatedly and on different days.