Background Crisis Standards of Care (CSC) ventilator triage guidance often includes prediction models such as the Sequential Organ Failure Assessment score (SOFA) to assess patients’ lCU survival. However, from Spring 2020, a number of publications and robust studies demonstrated inequitable outcomes across racial groups from using SOFA. For example, drawing on data of >100,000 patients, Ashana et al. 2021, showed that the metric erroneously overestimates the mortality of Black patients and wrongly excludes >9% from the highest prioritization. Removing SOFA’s creatinine sub-score reduced the miscalibration.

Objective To assess whether US state-level CSC included the SOFA score in January 2022, when Omicron led to 20 US states at >85% ICU bed capacity; if so, with or without measures aimed at reducing the risk of inequitable outcomes.

Methods We reviewed SOFA use in all US states’ publicly available CSC querying: a) the US Health Depts’ Technical Resources, Assistance Center, and Information Exchange (TRACIE) database; b) state health department websites; complemented with c) webbrowser searches. Documents were retrieved January 14-16, 2022. 4 team members independently retrieved and coded documents using a structured extraction tool (capturing date of issue; use of SOFA or other prediction models; measures proposed/noted to adjust SOFA for equity; and whether/to what extent remaining life expectancy was included in algorithms).

Results The study is ongoing at the time of abstract submission (min. 8 states currently use SOFA).

Discussion While influential commentators recently argued that CSC ‘cannot be expected to remedy historic and structural inequity’ and should merely ‘not exacerbate’ them (Hick et al. 2021), and while states continue to the use the SOFA score and fail to meet even this minimalist criterion, we need to better understand why robust evidence on inequitable outcomes is ignored, and what alternatives can be offered to avoid them in future heath emergencies.