2.2 | Statistics
The population was described by numbers/percentages for qualitative and
categorical variables and by averages (± standard deviation) for
quantitative variables. Quantitative data were compared between groups
by a student t-test or the Kruskal-Wallis test (if conditions of
validity, normality, and equality of variances were not present).
Comparison between two qualitative variables was performed via the
chi2 test or the Fisher’s exact test if applicable. A
multivariate analysis was conducted with a logistic regression model
(with a variable of a moderate to severe response versus a mild
response). The model considered an adjustment of significant factors in
the univariate situation and relevant parameters from a clinical point
of view in light of the literature. Interactions between factors were
tested. The difference tested was considered as significant at p
< 0.05. All analyses were carried out with IBM® SPSS®
Statistics.