Results (
Thai) 2:
[Copy]Copied!
characteristics of groups A and B were compared using the
unpaired t test or the Mann-Whitney U test for quantitative
variables. Depending on the groups and subsets selected for
comparisons (eg, music vs placebo-control, pre-music or
placebo-control [sham-MP3] vs post-music or placebo-control
[sham-MP3]), clinical and biological data were compared
using a Mann-Whitney U test (for unpaired data) or a
Wilcoxon matched-pairs signed-ranks test (for paired data); a
χ2 test (or the Fisher exact testwhen frequency was less than 5)
was selected for proportion comparisons. Comparisons
between VS parameters (pre- and post-) were performed
separately for morning and evening listening periods.
Additional post hoc comparative analyses were specifically
performed for subgroups of patients treated vs not treated with
corticosteroids, and those alive (survivors) vs in-hospital
deceased (non-survivors). Linear regression analysis was used
to demonstrate dynamic relationships between individual
blood changes in selected parameters. Sample size was
calculated for a 2-sample crossover designwithmusic listening
as the intervention, and a decreased narcotic per day
consumption (ie, as the selected reference standard) as a
post-listening change from baseline; for the primary endpoint.
Assuming an SD intragroup comparison of 1.875 mg/d and a
mean difference between music- and sham-MP3-listening
groups of 0.5mg/d, the sample size required to achieve an 80%
power at a 5% (α = .05) level of significance was n = 55 [14].
Being translated, please wait..
