Statistica 13.3 statistical package

For quantitative traits, distributions were checked. The mean and standard error of the mean (SEM), median (Me), and interquartile range (IQR) were calculated separately for patients tested at 0 months, 6 months, and the control group separately. Student’s t-tests for dependent samples and Wilcoxon’s paired rank test were used to compare the variables for binary studies. For comparison of COVID-19 men and women in the 2 separate studies, and COVID-19 patients studied at months 0 and 6 versus the control group, the Student’s t-test was used for independent samples and the Mann-Whitney, depending on their distribution. A statistical significance threshold of p < 0.05 was set. The analysis was performed with STATISTICA 13.3 statistical package, Polish version (STATSOFT, Krakow, Poland).

Counts (n) and percentages (%) attendance were counted for all qualitative parameters in the COVID-19 patient group. Three groups of patients with saturation were distinguished using the following cut-off values: ≤90%, ≤95%, and >95%. In the distinguished groups, the distributions of the variables were checked with the quantitative Shapiro–Wilk test. The mean ( $\overline{x}$ ) and standard error of the mean (SEM) were counted in the separate saturation groups. The difference between the mean and one-way analysis of variance (ANOVA) or Kruskal–Wallis test were used to assess differences between means. When significant differences were observed, the following tests were applied post-hoc: Bonferroni test or Dunn-Bonferroni test depending on the analysis applied. In order to determine the relationship between the parameter’s ocular and saturation, two types of correlation, Pearson or Spearman rank correlation, were considered. The following values were considered significant. The test values for which p ≤ 0.05 were considered significant. The statistical analysis was performed using the STATISTICA 13.3 statistical package (STATSOFT, PL, Kraków, Poland).