Measuring change in dry-mouth symptoms over time using the Xerostomia Inventory


W. Murray Thomson, Department of Oral Sciences, School of Dentistry, The University of Otago, Dunedin, New Zealand.
Tel.: 64 3 479 7116
Fax: 64 3 479 7113


Objective:  To examine the responsiveness to change of the Xerostomia Inventory (XI).

Background:  The XI is an 11-item summated rating scale which results in a single continuous scale score representing the severity of chronic xerostomia. While the XI has been used as an outcome measure in clinical research, the magnitude of a clinically meaningful change score has yet to be determined.

Methods:  This study comprises a secondary analysis of data from a longitudinal study of changes in xerostomia symptoms in two groups whose symptom trajectories were likely to differ substantially: the normal group was a convenience sample of asymptomatic middle-aged and older individuals with otherwise stable perceptions of mouth dryness; and the onset group comprised patients who were about to undergo radiotherapy for head/neck cancer (and would therefore be expected to develop more severe xerostomia after the baseline measurements). Statistical analyses examined cross-sectional construct validity and internal consistency, test-retest reliability and the measure's responsiveness and longitudinal construct validity. The mean change scores of those for whom ‘a little’ improvement was reported were used to determine the minimally important difference for the XI.

Results:  Over two-thirds of the onset group members reported dry mouth ‘frequently’ or ‘always’ at follow-up (2 months) and there was a concomitant increase in their mean XI score. Test-retest reliability was acceptable. Examination of within-individual change among those who changed and those for whom stability was observed, showed that only those who worsened had significantly greater XI scores at follow-up. The minimally important difference to reflect deterioration in xerostomia symptoms was determined to be 6 scale points. Between baseline and 2 months, the XI scores of 32 participants (33.7%) deteriorated by the minimally important difference (7.1% and 54.7% respectively among the normal and onset groups; p < 0.0001).

Conclusion:  The validity and responsiveness of the XI appear to be acceptable, and a change in XI score of 6 or more points is clinically meaningful.