2.5 Statistical analysis
Nominal variables are presented as absolute and relative (%)
frequencies. Continuous variables are presented in violin plots to show
frequency distribution, with median and quartiles. Data was evaluated
for significant statistical differences using i) the Chi-square test to
evaluate associations between two nominal variables, ii) the
non-parametric Kruskal-Wallis test for multiple comparisons and iii) the
non-parametric Mann-Whitney test for independent sample tests (Prism 8.0
for macOS; GraphPad Software, San Diego, CA, USA). Correlation was
assessed with the Spearman method. A local regression fitting line was
added to the scatter plot with a confidence level of 95%. Statistical
significance was set at p < 0.05 (* p < 0.05; ** p
< 0.01; *** p < 0.001; **** p < 0.0001).
Results