CoVA: An Acuity Score for Outpatient Screening that Predicts Coronavirus Disease 2019 Prognosis

Area of Science:

• Infectious Diseases
• Public Health
• Clinical Informatics

Background:

• Developing predictive tools for coronavirus disease 2019 (COVID-19) is crucial for managing patient outcomes in urgent care settings.
• Existing methods may not be readily automatable or validated for predicting severe COVID-19 outcomes.
• Urgent need for a reliable score to identify high-risk patients presenting for care.

Purpose of the Study:

• To develop and validate an automatable score to predict hospitalization, critical illness, or death in patients with COVID-19.
• To create a tool that can be easily implemented in urgent care settings.
• To improve risk stratification for patients with potential COVID-19.

Main Methods:

• The COVID-19 Acuity Score (CoVA) was developed using data from adult outpatients in a single-center study.
• Data were split into development and prospective cohorts for model training and validation.
• Outcomes included hospitalization, critical illness (ICU/ventilation), or death within 7 days; performance assessed by E/O and AUC.

Main Results:

• In prospective validation, CoVA demonstrated excellent performance for hospitalization (AUC: 0.76), critical illness (AUC: 0.79), and death (AUC: 0.93).
• The score showed good calibration with expected-to-observed event ratios near 1.0 for hospitalization and critical illness.
• Top predictors included age, diastolic blood pressure, oxygen saturation, COVID-19 testing status, and respiratory rate.

Conclusions:

• The COVID-19 Acuity Score (CoVA) is a prospectively validated, automatable tool for predicting adverse COVID-19 events.
• CoVA is suitable for use in the outpatient and urgent care settings.
• This score can aid clinicians in identifying patients at high risk for severe outcomes from COVID-19.