Fig 2.
Kaplan-Meier curves for survival without transfer to intensive care in unweighted sample (top panel) and sample used for inverse probability of treatment weighting (bottom panel). A multivariable logistic regression model was constructed to estimate each patient’s probability of receiving hydroxychloroquine given their baseline covariates (that is, the propensity score: variables in model included age, sex, and comorbidities). ICU=intensive care unit