Table 2. Association of COX inhibitors with COVID-19 severity.

For each combination of medication and disease indication, subcohorts were constructed using propensity matching. Ordinal logistic regression was used to assess association with COVID-19 severity. Odds ratios, s, 95% confidence intervals, and sample sizes (treated plus control) are shown.

^***p < 0.001,

^**p < 0.01,

^*p < 0.05 after Bonferroni correction for 21 tests. Details for each subcohort are provided in Supplemental Tables S5–S25 and Supplemental Figures S1–S21.