Statistical analysis
Normality of data was checked using the Shapiro-Wilk test. Demographic and baseline clinical characteristics, including comorbidities and blood parameters, were expressed as median (IQR) for continuous variables and as numbers (%) for categorical variables. Continuous variables were compared between tofacitinib and controls using the Mann-Whitney U-test test for two groups and the Kruskal Wallis test for three groups. Categorical variables across groups were compared by the Fisher’s exact test or Pearson’s χ² test.
All participants were followed up from the date of enrolment into the study until discharge or death. We conducted a survival analysis and compared the time to invasive mechanical ventilation or death between the treatment groups using unweighted Kaplan-Meier curves and univariable and multivariable Cox regression analysis with baseline fixed covariates. The effect of treatment was evaluated using an unadjusted and adjusted hazard ratio (HR) with 95% confidence interval (CI). The baseline model was adjusted for inverse propensity score weighting. Cox regression model validity was checked by evaluation of the proportional hazard assumption using Schoenfeld residuals.
We considered a two-sided p value test of less than 0.05 to be statistically significant. Statistical analyses were performed using IBM SPSS Statistics, version 22 (IBM Corporation, USA).