Data analysis
To identify potential differences between the distributions of the
COVID-19 drug treatment in non-post-COVID-19 syndrome and post-COVID-19
syndrome patients, Pearson’s chi-squared test was used in combination
with relative risks to measure the effect size in the univariate
analysis. Furthermore, using a binomial logistic regression, the
association between corticosteroids, antibiotics, or anticoagulants and
the development of the post-COVID-19 syndrome was determined. A p-value
of 0.05 or lower was considered to be significant. Missing data was
imputed using the expectation maximization method. A missing value of
5% or lower was considered to be acceptable for imputation.
Furthermore, data imputation was only necessary in one case. Statistics
were performed in R (version 4.0.3) and SPSS (version 28.0.1.0).