Skip to main content

Epidemiology, clinical course, and outcomes of critically ill adults with COVID-19 in New York City: a prospective cohort study

Our take —

The results of this study add to a growing body of literature suggesting that elevated levels of proinflammatory cytokines, such as IL-6, and markers of thrombosis, such as D-dimer, influence severe clinical presentation and mortality due to COVID-19, although there is no data presented on levels prior to infection, which limits attributing the elevations to infection. Additionally, similar to other studies, patients with cardiac and pulmonary comorbidities were at increased risk of mortality. Research on the pathogenesis of inflammatory and coagulation processes of COVID-19 are important and ongoing.

Study design

Prospective Cohort

Study population and setting

This prospective cohort study includes 257 critically ill adults with laboratory-confirmed COVID-19, who were admitted to two New York-Presbyterian hospitals between March 2 and April 1, 2020. Critical illness was defined as having acute hypoxaemic respiratory failure requiring mechanical ventilation or high-level supplemental oxygen at or during hospitalization. Patients were followed until April 28, 2020 or discharge from hospital.

Summary of Main Findings

Among 1150 adults diagnosed with COVID-19 at the two hospitals, 257 were critically ill and included in the study. Among those, 101 (39%) died in the hospital (median duration in hospital: 9 days), 94 (37%) remained hospitalized (median duration of hospitalization: 33 days), 4 (2%) were transferred to a different hospital, and 58 (23%) were discharged alive. In adjusted analyses, older age, chronic cardiac disease (coronary artery disease or congestive heart failure), COPD or interstitial lung disease, elevated IL-6 levels, and elevated D-dimer levels were independently associated with in-hospital mortality.

Study Strengths

Data on demographics, medical history, comorbidities, vital signs, and biochemical parameters were collected prospectively on standardized case report forms, which were developed in collaboration with WHO and the International Severe Acute Respiratory and Emerging Infection Consortium. Participants were followed for at least 28 days. The authors provide well-described justification for the variables included in their multivariable model.


Missing data were not imputed, so participants included in the multivariable model did not include the full study population, but the exact number is unclear. A large portion of participants remained hospitalized at the end of follow-up, and were (appropriately) censored, but their outcomes are unknown. The study population comprised patients from only 2 hospitals (both in New York City), which may limit generalizability. Data are limited to critically ill cases who have been hospitalized and findings may not be generalizable to less severe patients.

Value added

This is one of the largest prospective studies (to date) examining characteristics of critically ill COVID-19 patients in the United States.

This review was posted on: 17 June 2020