A Risk Score to Predict Admission to the Intensive Care Unit in Patients with COVID-19: the ABC-GOALS score

Abstract

Objective. To develop a score to predict the need for ICU admission in COVID-19. Methods. We assessed patients admitted to a COVID-19 center in Mexico. Patients were segregated into a group that required ICU admission, and a group that never required ICU admission. By logistic regression, we derived predictive models including clinical, laboratory, and imaging findings. The ABC-GOALS was constructed and compared to other scores. Results. We included 329 and 240 patients in the development and validation cohorts, respectively. One-hundred-fifteen patients from each cohort required ICU admission. The clinical (ABC-GOALSc), clinical+laboratory (ABC-GOALScl), clinical+laboratory+image (ABC-GOALSclx) models area under the curve were 0.79 (95%CI=0.74-0.83) and 0.77 (95%CI=0.71-0.83), 0.86 (95%CI=0.82-0.90) and 0.87 (95%CI=0.83-0.92), 0.88 (95%CI=0.84-0.92) and 0.86 (95%CI=0.81-0.90), in the development and validation cohorts, respectively. The ABC-GOALScl and ABC-GOALSclx outperformed other COVID-19 and pneumonia predictive scores. Conclusion. ABC-GOALS is a tool to timely predict the need for admission to ICU in COVID-19.

Keywords

• covid-19
• coronavirus
• pandemic
• intensive care
• prediction