Quantifying Utilitarian Outcomes to Inform Triage Ethics: Simulated Performance of a Ventilator Triage Protocol Under SARS-CoV-2 Pandemic Surge Conditions

Abstract

Background: Equitable protocols to triage scarce life-saving resources must be specified prior to disasters in order to promote transparency, trust and consistency. The degree to which proposed utilitarian protocols may maximize lives saved is unknown.Methods: We constructed a simulation model based on New York State 2015 guidelines for ventilator triage (NYS) compared to a first-come-first-served (FCFS) method for ventilator allocation under various hypothetical ventilator shortages. We included patients with laboratory confirmed SARs-CoV-2 infection admitted with respiratory failure requiring mechanical ventilation to three acute care hospitals in New York from 3/01/2020 and 5/27/2020. We estimated (1) survival rates, (2) number of excess deaths, (3) number of patients extubated early due to capacity constraints, (4) survival rates among patients not allocated a ventilator at triage or extubated early due to capacity constraints.Findings: 807 patients were included in the study. Survival rates were similar between the NYS and FCFS policies at all the survival probabilities estimated for those triaged to removal of the ventilator. At the lowest ventilator capacity, the NYS policy has an estimated survival of 28.5% (CI: 28.4-28.6), compared to 28.1% (CI: 27.7-28.5) for the FCFS policy. The two policies had similar estimated excess deaths above historical observed deaths.Interpretation: This simulation of a SOFA-based triage protocol for distributing ventilators during the COVID-19 pandemic revealed limitations in achieving the utilitarian goals these protocols are designed to fulfill. Quantifying these outcomes can inform a better balance competing moral aims.Funding: None to declare. Declaration of Interest: None to declare. Ethical Approval: This study was approved by the Albert Einstein College of Medicine/Montefiore Medical Center IRB (IRB # 2020-12128) and the Columbia University IRB (IRB # AAAT4062).