Fig. 2.

Association between metformin usage and COVID-19 severity in PCOS and prediabetes cohorts. Odds ratios (x-axis) and adjusted p-values are shown for the given combination of drug, cohort and outcome. For the PCOS cohort, there were insufficient patients with severe or worse disease or mortality/hospice for a logistic regression to be performed. All p-values are Bonferonni corrected. A comparison of the estimates obtained before and after cohort weighting provides an estimate of the effect size (Supplementary Fig. S5). Note that estimates are comparable, even for groups that show large effect sizes according to Cohen’s D (Table 2).