Timing of Invasive Mechanical Ventilation and Death in Critically Ill Adults With COVID-19: A Multicenter Cohort Study

Adam Green , Jean-Sebastien Rachoin , Christa Schorr, Phil Dellinger, Jonathan D. Casey, Isabel Park, Shruti Gupta, Rebecca M. Baron, Shahzad Shaefi, Krystal Hunter, David E. Leaf ,for the STOP-COVID Investigators


To investigate if the timing of initiation of invasive mechanical ventilation (IMV) for critically ill patients with COVID-19 is associated with mortality.


Critically ill patients with coronavirus disease 2019 (COVID-19) who receive invasive mechanical ventilation (IMV) have in-hospital mortality rates as high as 40–50% [1, 2]. Despite the development of anti-inflammatory [3], immunomodulatory [4], and antiviral agents [5] to treat severe illness from COVID-19, mortality in this patient population remains extremely high. Additional therapeutic strategies are therefore urgently needed.

Materials and Methods

We used data from the Study of the Treatment and Outcomes in Critically ill Patients with COVID-19 (STOP-COVID), a multicenter cohort study that enrolled consecutive adults with laboratory-confirmed COVID-19 admitted to participating ICUs at 68 hospitals across the United States. Study personnel at each site collected data by detailed chart review and used a standardized electronic case report form (Research Electronic Data Capture [REDCap]) to enter data into a secure online database.


The median follow-up for the patients who initiated IMV early and late was 18 (IQR, 10–31) and 19 (IQR, 9–30) days, respectively (overall, 18 [IQR, 10–31] days). A total of 824 patients (43.9%) died, 1035 (55.1%) were discharged alive, and 20 (1.1%) remained hospitalized at day 90. The 824 patients who died included 644 of the 1526 (42.2%) who initiated IMV early and 180 of the 353 (51.0%) who initiated IMV late (unadjusted HR, 0.80 [95% CI, 0.67–0.94]; Fig 2). Causes of death are shown in S1 Table in S1 File.


Our study has several strengths. First, in contrast to the studies included in the meta-analysis described above, we designed our inclusion and exclusion criteria to be representative of those that would be used in a hypothetical RCT of early versus delayed initiation of IMV, excluding patients who would be ineligible to participate in such a trial (e.g., those with severe hypoxemia [PaO2:FiO2 ratio <100 mm Hg] or refractory shock on ICU admission). Second, we used analytic approaches to prevent immortal time bias and to comprehensively adjust for confounding, leveraging the detailed and longitudinal severity-of-illness data collected in STOP-COVID. Third, whereas most prior studies were single-center, ours included data from multiple geographically-diverse hospitals across the United States, thereby increasing the generalizability of our findings. Fourth, our results were consistent across multiple sensitivity and subgroup analyses.


In summary, we found that patients who initiated IMV early had a considerably lower adjusted risk of death compared to those who initiated IMV late. Future studies, ideally in the form of carefully designed and adequately powered RCTs, are needed to identify optimal therapeutic strategies surrounding IMV in critically ill patients with COVID-19 to reduce the very high mortality in this population.

Citation: Green A, Rachoin J-S, Schorr C, Dellinger P, Casey JD, Park I, et al. (2023) Timing of invasive mechanical ventilation and death in critically ill adults with COVID-19: A multicenter cohort study. PLoS ONE 18(6): e0285748.

Editor: Luis Felipe Reyes, Universidad de La Sabana, COLOMBIA

Received: November 16, 2022; Accepted: May 2, 2023; Published: June 28, 2023

Copyright: © 2023 Green et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The authors are not able to share the raw dataset publicly for both legal and ethical reasons. The STOP-COVID Coordinating Center at Brigham and Women’s Hospital has a separate data use agreement (DUA) with each of the 67 other institutions that contributed data to the study. Each DUA has its own specifications to comply with local IRB policies from each of the contributing sites regarding sharing of data. The DUAs do not give the STOP-COVID investigators permission to make the data publicly available in any form (with our without identifiers). Data are available upon request from Joseph Niola, Senior Administrator, Division of Renal Medicine, Brigham and Women’s Hospital, via email ([email protected]) for researchers who meet the criteria for access to confidential data.

Funding: The authors received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Harvard Medical School - Leadership in Medicine Southeast Asia47th IHF World Hospital Congress