Performance of crisis standards of care guidelines in a cohort of critically ill COVID-19 patients in the United States

Abstract

Many US states published crisis standards of care (CSC) guidelines for allocating scarce critical care resources during the COVID-19 pandemic. However, the performance of these guidelines in maximizing their population benefit has not been well tested. In 2,272 adults with COVID-19 requiring mechanical ventilation drawn from the Study of the Treatment and Outcomes in Critically Ill Patients with COVID-19 (STOP-COVID) multicenter cohort, we test the following three approaches to CSC algorithms: Sequential Organ Failure Assessment (SOFA) scores grouped into ranges, SOFA score ranges plus comorbidities, and a hypothetical approach using raw SOFA scores not grouped into ranges. We find that area under receiver operating characteristic (AUROC) curves for all three algorithms demonstrate only modest discrimination for 28-day mortality. Adding comorbidity scoring modestly improves algorithm performance over SOFA scores alone. The algorithm incorporating comorbidities has modestly worse predictive performance for Black compared to white patients. CSC algorithms should be empirically examined to refine approaches to the allocation of scarce resources during pandemics and to avoid potential exacerbation of racial inequities.

Keywords: ARDS; COVID-19; acute respiratory distress syndrome; crisis standards of care; critical care; intensive care; medical ethics; triage.

Conflict of interest statement

The authors have no conflicts of interest relevant to this work. For unrelated work, the authors have the following disclosures: S.P.K. served as a consultant for Resolve to Save Lives and has previously led a partnership on multiple chronic conditions at the Icahn School of Medicine at Mount Sinai with Teva Pharmaceuticals. W.B.F. serves as a consultant for Alosa Health and Aetion. He also received an honorarium for a presentation to Blue Cross Blue Shield of Massachusetts. E.Y.K. is a co-investigator in NCT04389671 (Windtree Therapeutics) testing lucinactant (surfactant-like treatment) in COVID-19 patients and received unrelated research funding from Bayer AG, N.I.H., the American Heart Association, the American Lung Association, and the American Thoracic Society.

© 2021 The Author(s).

Figures

Graphical abstract

Figure 1

Association of priority score or category with 28-day mortality (A–C) The number of patients who survived or died at 28 days after ICU admission and intubation are shown for each priority point value (or category) for each algorithm. (A) New York (SOFA score groups only). (B) Colorado (SOFA score groups and comorbidities). (C) Hypothetical algorithm of raw (ungrouped) SOFA score. (D and E) AUROC curves for discrimination of 28-day mortality by priority scores are shown for the following algorithms: New York (SOFA score groups) (D), Colorado (SOFA score groups and comorbidities) (D), and raw SOFA scores. (E) Colorado SOFA score component, Colorado comorbidity scoring component, or full Colorado algorithm (SOFA and comorbidities).