Four patients did not complete the research. They did not
appear in the interviews following the first interview and the reason
for this could not be ascertained because they could not be contacted. The analyses were performed on the data from the
remaining 30 patients. Post hoc sample size calculation was based
on the main outcome variable, HAM-D scores. Taking alpha error as
0.05, and effect size as 1.1; 16 participants in non-chewers group
and 14 in chewers group were needed to reach a power of 82.6%.