ABSTRACT
A clinical interpretation of the Randomized Evaluation of COVID-19 Therapy (RECOVERY) study was performed to provide a useful tool to understand whether, when, and to whom dexamethasone should be administered during hospitalization for COVID-19. A post hoc analysis of data published in the preliminary report of the RECOVERY study was performed to calculate the person-based number needed to treat (NNT) and number needed to harm (NNH) of 6 mg dexamethasone once daily for up to 10 days vs. usual care with respect to mortality. At day 28, the NNT of dexamethasone vs. usual care was 36.0 (95%CI 24.9-65.1, p < 0.05) in all patients, 8.3 (95%CI 6.0-13.1, p < 0.05) in patients receiving invasive mechanical ventilation, and 34.6 (95%CI 22.1-79.0, p < 0.05) in patients receiving oxygen only (with or without noninvasive ventilation). Dexamethasone increased mortality compared with usual care in patients not requiring oxygen supplementation, leading to a NNH value of 26.7 (95%CI 18.1-50.9, p < 0.05). NNT of dexamethasone vs. usual care was 17.3 (95%CI 14.9-20.6) in subjects <70 years, 27.0 (95%CI 18.5-49.8) in men, and 16.2 (95%CI 13.2-20.8) in patients in which the onset of symptoms was >7 days. Dexamethasone is effective in male subjects < 70 years that require invasive mechanical ventilation experiencing symptoms from >7 days and those patients receiving oxygen without invasive mechanical ventilation; it should be avoided in patients not requiring respiratory support.