Evaluation of the effectiveness of remdesivir in severe COVID-19 using observational data from a prospective national cohort study

Background Remdesivir has been evaluated in clinical trial populations, but there is a sparsity of evidence evaluating effectiveness in general populations. Methods Adults eligible to be treated with remdesivir, requiring oxygen but not ventilated, were identified from UK patients hospitalised with COVID-19. Patients treated with remdesivir within 24h of hospitalisation were compared with propensity-score matched controls; estimates of effectiveness were calculated for short-term outcomes (14-day mortality, 28-day mortality, time-to-recovery among others) using multivariable modelling. Results 9,278 out of 39,330 patients satisfied eligibility criteria. 1,549 patients were identified as ‘treated’ and matched with 4,964 controls. Patients were 62% male, mean (SD) age 63.1 (15.6) years, 80% ‘White’ ethnicity, and symptomatic for a median of 6 days prior to baseline. There was no statistically significant benefit of remdesivir at 14 days in terms of mortality or clinical status; there were signals of effectiveness in time-to-recovery after day 9, and a reduction in 28-day mortality. Conclusion In a real-world setting, initiation of remdesivir within 24h of hospitalisation in conjunction with standard of care was not associated with a benefit at 14 days but supports clinical trial evidence of a potential reduction in 28-day mortality