Resource Allocation & Triage Models

This article by Chelen et. al. reports an assessment of policy preparedness and substantive triage criteria for ventilator allocation based on a survey from March-April 2020, in the early initial surge of cases in the U.S. Nearly all of the surveyed organizations were in the process of developing allocation policies, but only half had already adopted formal policies. Most policies used objective criteria, specified inapplicable criteria, separated triage and clinical decision making, provided detailed reassessment plans, offered an appeals process, and addressed palliative care. All one policy referred to sequential organ failure assessment (SOFA) score as a triage criterion, and several included categorial exclusion criteria. The article provides a detailed breakdown of the allocation plans, including commonalities and key differences across factors such as criteria used, exclusion criteria, decision-making authority, appeals, documentation and communication, and reassessment.

This article by Kesler et al. reports a study of the resource allocation guidance developed in Minnesota during the pandemic. The plan was based on existed guidance created during the 2009 Influenza pandemic and focused on ICU and ventilator rationing under Crisis Standard of Care (CSC) conditions. Like many such plans, the primary aim of the guidance was to save the most lives to maximize population benefit, and SOFA score was the primary foundation of the scoring system for patient triage. However, the group that developed the Minnesota guidance added two additional factors (comorbidities likely to impact survival and anticipated need for ventilator) into the Ventilator Allocation Score (VAS) to address limitations of SOFA score for this purpose.

• This study assessed how accurately the VAS grouped patients into priority tiers, how many patients were placed into each tier, and whether there was any evidence of bias against any demographic group.
• The study found that VAS was able to identify patients with very high rates of short-term mortality, but it was not able to distinguish between patients with lower scores and it failed to appropriately group patients into priority tiers. The authors conclude that the utility of the system as a triage tool is therefore limited.
• The authors also conclude that despite adherence to ethical and legal guidance and intention to treat people fairly and maximize population benefit, the scoring system may not have achieved either objective. Among other problems, scores may have been biased against BIPOC patients, potentially perpetuating or exacerbating prevailing inequities despite intention to the contrary.

Closely related, this article by Miller et al. investigates whether reliance on SOFA scores is associated with bias against Black patients in CSC. While this study was conducted during the pandemic, the data used were from 2014-2015 ICU admissions. The study found that while median SOFA scores were not statistically significantly different between Black and white patients, mortality was lower among Black patients compared to white patients with equivalent SOFA scores. This means that SOFA sore was associated with overestimating mortality among Black patients compared to white patients. This creates a structural disadvantage in the event of scarce resource allocation because it means that Black patients would be more likely to be deprioritized for resources due to systematic misclassification into lower priority tiers. The authors conclude that allocation guidelines should be revised to correct this inequity and that alternative methods should be developed for more equitable triage.

Racial/Ethnic Equity & Mortality

This article by Asch et al. reports a study of COVID-19 patient mortality that found that higher mortality rates among Black patients compared to white patients were associated with the hospitals at which Black patients disproportionately received care.

• This study adds to a considerable body of literature on the disproportionate impact of the pandemic on Black communities, among other racial and ethnic minority populations in the U.S. Throughout the pandemic, Black people have suffered higher rates of infection, hospitalization, and mortality due to a variety of structural factors.
• The outcome of interest was 30-day in-hospital mortality or discharge to hospice among patients admitted to the hospital with a diagnosis of COVID-19.
• The study included only Medicare Advantage enrollees, reducing variation on the basis of insurance, and also controlled for clinical, sociodemographic, and other factors (age, sex, zip code-level income, comorbidities, admission from nursing facility, and date of admission).
• After controlling for those variables, there was an association between mortality and the hospitals where care was received, which supports the conclusion that differences in mortality did not reflect differences in patient characteristics, but rather characteristics of the facilities where care was received. This suggests that if Black patients received care at the same hospitals as white patients, mortality rates would likely be similar between the two populations, reflecting an additional structural factor affecting the health of Black communities.