Statistical Analysis
Pilot
study:
Comprehensibility was calculated as the percentage of patients who
stated that the instrument was understandable. Feasibility was assessed
by examining the number of missing responses on the returned
questionnaires. Internal consistency analysis was performed with the
Cronbach α coefficient.
A content validity index (CVI) based on expert ratings of relevance was
used to quantify content validity. We calculated the CVI for each item
(I-CVI) and for the overall scale (OS-CVI). The I-CVI was calculated as
the number of experts giving a rating of either 3 or 4, divided by the
number of experts. To determine OS-CVI, we calculated the I-CVI of each
item on the questionnaire and then computed the average I-CVI across
items[18] (Supplementary Table 2).
Validation process: Feasibility was assessed by examining the number of
missing responses on the returned questionnaires and the time required
to complete the questionnaire in minutes.A reliability analysis was
conducted, following the test-retest procedure. Internal consistency
analysis was performed with the Cronbach α coefficient. Construct
validity was assessed using known groups validity and
convergent/divergent validity. The Mann-Whitney U test was used to test
for statistically significant differences between two subgroups of
frequency (fewer than 2 episodes and more than 4 episodes per week).
Convergent/divergent validity was assessed by correlating CSOC scores
with SF-36 and VAS using the Pearson correlation coefficient.
Responsiveness was evaluated comparing CSOC results before and after
sialendoscopy by paired sample t-test.