Figure 1. Association Between Anti-CD20 Therapies and Severe COVID-19 in All Patients With Relapsing-Remitting Multiple Sclerosis (RRMS) and by Subgroup.

Observed numbers of events are reported in the nonweighted population without imputation of missing data. Percentages and odds ratios (ORs) were estimated in the weighted population with imputation of missing data (including outcomes). The analysis was propensity score weighted (average treatment effect estimand). Variables used in the propensity score estimation were age, sex, time since multiple sclerosis onset, Expanded Disability Status Scale (EDSS) score, comorbidities (cardiovascular disease, pulmonary disease, diabetes, obesity, and active tobacco smoking), exposure to high-dose methylprednisolone in the month before COVID-19, postvaccine COVID-19 (infection occurring at least 7 days after the second dose of COVID-19 vaccine), previous COVID-19, COVID-19 variant (original Alpha, Delta, or Omicron), and treatment of COVID-19 with monoclonal antibody. Severe COVID-19 was defined as hospitalization with any mode of oxygenation or death. Age was divided into 2 groups based on the median for patients with RRMS, and subgroup analyses were conducted in age categories defined accordingly. P values for the subgroup analyses are for the interaction. Squares indicate ORs, with horizontal lines indicating 95% CIs.