For a study, researchers sought to determine the time to recovery of command-following and the relationships between hypoxemia and recovery time. Using Kaplan Meier cumulative-incidence curves and Cox proportional hazard curves, they estimated the duration from intubation to recovery of command-following in this multicenter, retrospective, cohort study during the initial spike of the United States pandemic (March–July 2020). Patients were included if they were admitted to 1 of 3 hospitals due to severe coronavirus disease 2019 (COVID-19), required endotracheal intubation for at least 7 days, and had awareness impairment (Glasgow Coma Scale motor score<6). About 571 of the 795 patients recovered command-following. The median duration for command-following to recover was 30 days (95% CI = 27–32 days). For patients with at least 1 incident of arterial partial pressure of oxygen (PaO2) value of less than or equal to 55 mmHg, the median time to recovery of command-following increased by 16 days (P<0.001), and 25% recovered in more than or equal to 10 days after discontinuation of mechanical ventilation. The time to recovery of command-following was associated with hypoxemia (PaO2≤55 mmHg hazard ratio [HR]=0.56, 95% CI=0.46–0.68; PaO2≤70 HR=0.88, 95% CI=0.85–0.91), and each additional day of hypoxemia decreased the likelihood of recovery, controlling for confounders such as sedation. These findings were validated by patients without any imaging indications of structural brain injury (n=209) and a non-overlapping second surge group (n=427, October 2020 to April 2021). Survivors of severe COVID-19 typically regain consciousness weeks after mechanical breathing is discontinued. Hypoxemia was connected with more severe recovery times. This relationship cannot be explained by sedation or brain injury identified by clinical imaging and should be considered when deciding on life-sustaining medications.